T3Q is a fucking cunt

* Add KVM feature extensions for NXT core

virt/kvm/arm/aarch32.c

@@ -0,0 +1,232 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * (not much of an) Emulation layer for 32bit guests.
+ *
+ * Copyright (C) 2012,2013 - ARM Ltd
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ *
+ * based on arch/arm/kvm/emulate.c
+ * Copyright (C) 2012 - Virtual Open Systems and Columbia University
+ * Author: Christoffer Dall <c.dall@virtualopensystems.com>
+ */
+
+#include <linux/bits.h>
+#include <linux/kvm_host.h>
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_hyp.h>
+
+#define DFSR_FSC_EXTABT_LPAE	0x10
+#define DFSR_FSC_EXTABT_nLPAE	0x08
+#define DFSR_LPAE		BIT(9)
+
+/*
+ * Table taken from ARMv8 ARM DDI0487B-B, table G1-10.
+ */
+static const u8 return_offsets[8][2] = {
+	[0] = { 0, 0 },		/* Reset, unused */
+	[1] = { 4, 2 },		/* Undefined */
+	[2] = { 0, 0 },		/* SVC, unused */
+	[3] = { 4, 4 },		/* Prefetch abort */
+	[4] = { 8, 8 },		/* Data abort */
+	[5] = { 0, 0 },		/* HVC, unused */
+	[6] = { 4, 4 },		/* IRQ, unused */
+	[7] = { 4, 4 },		/* FIQ, unused */
+};
+
+static bool pre_fault_synchronize(struct kvm_vcpu *vcpu)
+{
+	preempt_disable();
+	if (kvm_arm_vcpu_loaded(vcpu)) {
+		kvm_arch_vcpu_put(vcpu);
+		return true;
+	}
+
+	preempt_enable();
+	return false;
+}
+
+static void post_fault_synchronize(struct kvm_vcpu *vcpu, bool loaded)
+{
+	if (loaded) {
+		kvm_arch_vcpu_load(vcpu, smp_processor_id());
+		preempt_enable();
+	}
+}
+
+/*
+ * When an exception is taken, most CPSR fields are left unchanged in the
+ * handler. However, some are explicitly overridden (e.g. M[4:0]).
+ *
+ * The SPSR/SPSR_ELx layouts differ, and the below is intended to work with
+ * either format. Note: SPSR.J bit doesn't exist in SPSR_ELx, but this bit was
+ * obsoleted by the ARMv7 virtualization extensions and is RES0.
+ *
+ * For the SPSR layout seen from AArch32, see:
+ * - ARM DDI 0406C.d, page B1-1148
+ * - ARM DDI 0487E.a, page G8-6264
+ *
+ * For the SPSR_ELx layout for AArch32 seen from AArch64, see:
+ * - ARM DDI 0487E.a, page C5-426
+ *
+ * Here we manipulate the fields in order of the AArch32 SPSR_ELx layout, from
+ * MSB to LSB.
+ */
+static unsigned long get_except32_cpsr(struct kvm_vcpu *vcpu, u32 mode)
+{
+	u32 sctlr = vcpu_cp15(vcpu, c1_SCTLR);
+	unsigned long old, new;
+
+	old = *vcpu_cpsr(vcpu);
+	new = 0;
+
+	new |= (old & PSR_AA32_N_BIT);
+	new |= (old & PSR_AA32_Z_BIT);
+	new |= (old & PSR_AA32_C_BIT);
+	new |= (old & PSR_AA32_V_BIT);
+	new |= (old & PSR_AA32_Q_BIT);
+
+	// CPSR.IT[7:0] are set to zero upon any exception
+	// See ARM DDI 0487E.a, section G1.12.3
+	// See ARM DDI 0406C.d, section B1.8.3
+
+	new |= (old & PSR_AA32_DIT_BIT);
+
+	// CPSR.SSBS is set to SCTLR.DSSBS upon any exception
+	// See ARM DDI 0487E.a, page G8-6244
+	if (sctlr & BIT(31))
+		new |= PSR_AA32_SSBS_BIT;
+
+	// CPSR.PAN is unchanged unless SCTLR.SPAN == 0b0
+	// SCTLR.SPAN is RES1 when ARMv8.1-PAN is not implemented
+	// See ARM DDI 0487E.a, page G8-6246
+	new |= (old & PSR_AA32_PAN_BIT);
+	if (!(sctlr & BIT(23)))
+		new |= PSR_AA32_PAN_BIT;
+
+	// SS does not exist in AArch32, so ignore
+
+	// CPSR.IL is set to zero upon any exception
+	// See ARM DDI 0487E.a, page G1-5527
+
+	new |= (old & PSR_AA32_GE_MASK);
+
+	// CPSR.IT[7:0] are set to zero upon any exception
+	// See prior comment above
+
+	// CPSR.E is set to SCTLR.EE upon any exception
+	// See ARM DDI 0487E.a, page G8-6245
+	// See ARM DDI 0406C.d, page B4-1701
+	if (sctlr & BIT(25))
+		new |= PSR_AA32_E_BIT;
+
+	// CPSR.A is unchanged upon an exception to Undefined, Supervisor
+	// CPSR.A is set upon an exception to other modes
+	// See ARM DDI 0487E.a, pages G1-5515 to G1-5516
+	// See ARM DDI 0406C.d, page B1-1182
+	new |= (old & PSR_AA32_A_BIT);
+	if (mode != PSR_AA32_MODE_UND && mode != PSR_AA32_MODE_SVC)
+		new |= PSR_AA32_A_BIT;
+
+	// CPSR.I is set upon any exception
+	// See ARM DDI 0487E.a, pages G1-5515 to G1-5516
+	// See ARM DDI 0406C.d, page B1-1182
+	new |= PSR_AA32_I_BIT;
+
+	// CPSR.F is set upon an exception to FIQ
+	// CPSR.F is unchanged upon an exception to other modes
+	// See ARM DDI 0487E.a, pages G1-5515 to G1-5516
+	// See ARM DDI 0406C.d, page B1-1182
+	new |= (old & PSR_AA32_F_BIT);
+	if (mode == PSR_AA32_MODE_FIQ)
+		new |= PSR_AA32_F_BIT;
+
+	// CPSR.T is set to SCTLR.TE upon any exception
+	// See ARM DDI 0487E.a, page G8-5514
+	// See ARM DDI 0406C.d, page B1-1181
+	if (sctlr & BIT(30))
+		new |= PSR_AA32_T_BIT;
+
+	new |= mode;
+
+	return new;
+}
+
+static void prepare_fault32(struct kvm_vcpu *vcpu, u32 mode, u32 vect_offset)
+{
+	unsigned long spsr = *vcpu_cpsr(vcpu);
+	bool is_thumb = (spsr & PSR_AA32_T_BIT);
+	u32 return_offset = return_offsets[vect_offset >> 2][is_thumb];
+	u32 sctlr = vcpu_cp15(vcpu, c1_SCTLR);
+
+	*vcpu_cpsr(vcpu) = get_except32_cpsr(vcpu, mode);
+
+	/* Note: These now point to the banked copies */
+	vcpu_write_spsr(vcpu, host_spsr_to_spsr32(spsr));
+	*vcpu_reg32(vcpu, 14) = *vcpu_pc(vcpu) + return_offset;
+
+	/* Branch to exception vector */
+	if (sctlr & (1 << 13))
+		vect_offset += 0xffff0000;
+	else /* always have security exceptions */
+		vect_offset += vcpu_cp15(vcpu, c12_VBAR);
+
+	*vcpu_pc(vcpu) = vect_offset;
+}
+
+void kvm_inject_undef32(struct kvm_vcpu *vcpu)
+{
+	bool loaded = pre_fault_synchronize(vcpu);
+
+	prepare_fault32(vcpu, PSR_AA32_MODE_UND, 4);
+	post_fault_synchronize(vcpu, loaded);
+}
+
+/*
+ * Modelled after TakeDataAbortException() and TakePrefetchAbortException
+ * pseudocode.
+ */
+static void inject_abt32(struct kvm_vcpu *vcpu, bool is_pabt,
+			 unsigned long addr)
+{
+	u32 vect_offset;
+	u32 *far, *fsr;
+	bool is_lpae;
+	bool loaded;
+
+	loaded = pre_fault_synchronize(vcpu);
+
+	if (is_pabt) {
+		vect_offset = 12;
+		far = &vcpu_cp15(vcpu, c6_IFAR);
+		fsr = &vcpu_cp15(vcpu, c5_IFSR);
+	} else { /* !iabt */
+		vect_offset = 16;
+		far = &vcpu_cp15(vcpu, c6_DFAR);
+		fsr = &vcpu_cp15(vcpu, c5_DFSR);
+	}
+
+	prepare_fault32(vcpu, PSR_AA32_MODE_ABT, vect_offset);
+
+	*far = addr;
+
+	/* Give the guest an IMPLEMENTATION DEFINED exception */
+	is_lpae = (vcpu_cp15(vcpu, c2_TTBCR) >> 31);
+	if (is_lpae) {
+		*fsr = DFSR_LPAE | DFSR_FSC_EXTABT_LPAE;
+	} else {
+		/* no need to shuffle FS[4] into DFSR[10] as its 0 */
+		*fsr = DFSR_FSC_EXTABT_nLPAE;
+	}
+
+	post_fault_synchronize(vcpu, loaded);
+}
+
+void kvm_inject_dabt32(struct kvm_vcpu *vcpu, unsigned long addr)
+{
+	inject_abt32(vcpu, false, addr);
+}
+
+void kvm_inject_pabt32(struct kvm_vcpu *vcpu, unsigned long addr)
+{
+	inject_abt32(vcpu, true, addr);
+}

virt/kvm/arm/hyp/aarch32.c

@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Hyp portion of the (not much of an) Emulation layer for 32bit guests.
+ *
+ * Copyright (C) 2012,2013 - ARM Ltd
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ *
+ * based on arch/arm/kvm/emulate.c
+ * Copyright (C) 2012 - Virtual Open Systems and Columbia University
+ * Author: Christoffer Dall <c.dall@virtualopensystems.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_hyp.h>
+
+/*
+ * stolen from arch/arm/kernel/opcodes.c
+ *
+ * condition code lookup table
+ * index into the table is test code: EQ, NE, ... LT, GT, AL, NV
+ *
+ * bit position in short is condition code: NZCV
+ */
+static const unsigned short cc_map[16] = {
+	0xF0F0,			/* EQ == Z set            */
+	0x0F0F,			/* NE                     */
+	0xCCCC,			/* CS == C set            */
+	0x3333,			/* CC                     */
+	0xFF00,			/* MI == N set            */
+	0x00FF,			/* PL                     */
+	0xAAAA,			/* VS == V set            */
+	0x5555,			/* VC                     */
+	0x0C0C,			/* HI == C set && Z clear */
+	0xF3F3,			/* LS == C clear || Z set */
+	0xAA55,			/* GE == (N==V)           */
+	0x55AA,			/* LT == (N!=V)           */
+	0x0A05,			/* GT == (!Z && (N==V))   */
+	0xF5FA,			/* LE == (Z || (N!=V))    */
+	0xFFFF,			/* AL always              */
+	0			/* NV                     */
+};
+
+/*
+ * Check if a trapped instruction should have been executed or not.
+ */
+bool __hyp_text kvm_condition_valid32(const struct kvm_vcpu *vcpu)
+{
+	unsigned long cpsr;
+	u32 cpsr_cond;
+	int cond;
+
+	/* Top two bits non-zero?  Unconditional. */
+	if (kvm_vcpu_get_hsr(vcpu) >> 30)
+		return true;
+
+	/* Is condition field valid? */
+	cond = kvm_vcpu_get_condition(vcpu);
+	if (cond == 0xE)
+		return true;
+
+	cpsr = *vcpu_cpsr(vcpu);
+
+	if (cond < 0) {
+		/* This can happen in Thumb mode: examine IT state. */
+		unsigned long it;
+
+		it = ((cpsr >> 8) & 0xFC) | ((cpsr >> 25) & 0x3);
+
+		/* it == 0 => unconditional. */
+		if (it == 0)
+			return true;
+
+		/* The cond for this insn works out as the top 4 bits. */
+		cond = (it >> 4);
+	}
+
+	cpsr_cond = cpsr >> 28;
+
+	if (!((cc_map[cond] >> cpsr_cond) & 1))
+		return false;
+
+	return true;
+}
+
+/**
+ * adjust_itstate - adjust ITSTATE when emulating instructions in IT-block
+ * @vcpu:	The VCPU pointer
+ *
+ * When exceptions occur while instructions are executed in Thumb IF-THEN
+ * blocks, the ITSTATE field of the CPSR is not advanced (updated), so we have
+ * to do this little bit of work manually. The fields map like this:
+ *
+ * IT[7:0] -> CPSR[26:25],CPSR[15:10]
+ */
+static void __hyp_text kvm_adjust_itstate(struct kvm_vcpu *vcpu)
+{
+	unsigned long itbits, cond;
+	unsigned long cpsr = *vcpu_cpsr(vcpu);
+	bool is_arm = !(cpsr & PSR_AA32_T_BIT);
+
+	if (is_arm || !(cpsr & PSR_AA32_IT_MASK))
+		return;
+
+	cond = (cpsr & 0xe000) >> 13;
+	itbits = (cpsr & 0x1c00) >> (10 - 2);
+	itbits |= (cpsr & (0x3 << 25)) >> 25;
+
+	/* Perform ITAdvance (see page A2-52 in ARM DDI 0406C) */
+	if ((itbits & 0x7) == 0)
+		itbits = cond = 0;
+	else
+		itbits = (itbits << 1) & 0x1f;
+
+	cpsr &= ~PSR_AA32_IT_MASK;
+	cpsr |= cond << 13;
+	cpsr |= (itbits & 0x1c) << (10 - 2);
+	cpsr |= (itbits & 0x3) << 25;
+	*vcpu_cpsr(vcpu) = cpsr;
+}
+
+/**
+ * kvm_skip_instr - skip a trapped instruction and proceed to the next
+ * @vcpu: The vcpu pointer
+ */
+void __hyp_text kvm_skip_instr32(struct kvm_vcpu *vcpu, bool is_wide_instr)
+{
+	u32 pc = *vcpu_pc(vcpu);
+	bool is_thumb;
+
+	is_thumb = !!(*vcpu_cpsr(vcpu) & PSR_AA32_T_BIT);
+	if (is_thumb && !is_wide_instr)
+		pc += 2;
+	else
+		pc += 4;
+
+	*vcpu_pc(vcpu) = pc;
+
+	kvm_adjust_itstate(vcpu);
+}

virt/kvm/arm/hyp/timer-sr.c

@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2012-2015 - ARM Ltd
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ */
+
+#include <clocksource/arm_arch_timer.h>
+#include <linux/compiler.h>
+#include <linux/kvm_host.h>
+
+#include <asm/kvm_hyp.h>
+
+void __hyp_text __kvm_timer_set_cntvoff(u32 cntvoff_low, u32 cntvoff_high)
+{
+	u64 cntvoff = (u64)cntvoff_high << 32 | cntvoff_low;
+	write_sysreg(cntvoff, cntvoff_el2);
+}
+
+/*
+ * Should only be called on non-VHE systems.
+ * VHE systems use EL2 timers and configure EL1 timers in kvm_timer_init_vhe().
+ */
+void __hyp_text __timer_disable_traps(struct kvm_vcpu *vcpu)
+{
+	u64 val;
+
+	/* Allow physical timer/counter access for the host */
+	val = read_sysreg(cnthctl_el2);
+	val |= CNTHCTL_EL1PCTEN | CNTHCTL_EL1PCEN;
+	write_sysreg(val, cnthctl_el2);
+}
+
+/*
+ * Should only be called on non-VHE systems.
+ * VHE systems use EL2 timers and configure EL1 timers in kvm_timer_init_vhe().
+ */
+void __hyp_text __timer_enable_traps(struct kvm_vcpu *vcpu)
+{
+	u64 val;
+
+	/*
+	 * Disallow physical timer access for the guest
+	 * Physical counter access is allowed
+	 */
+	val = read_sysreg(cnthctl_el2);
+	val &= ~CNTHCTL_EL1PCEN;
+	val |= CNTHCTL_EL1PCTEN;
+	write_sysreg(val, cnthctl_el2);
+}

virt/kvm/arm/mmio.c

@@ -0,0 +1,219 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2012 - Virtual Open Systems and Columbia University
+ * Author: Christoffer Dall <c.dall@virtualopensystems.com>
+ */
+
+#include <linux/kvm_host.h>
+#include <asm/kvm_mmio.h>
+#include <asm/kvm_emulate.h>
+#include <trace/events/kvm.h>
+
+#include "trace.h"
+
+void kvm_mmio_write_buf(void *buf, unsigned int len, unsigned long data)
+{
+	void *datap = NULL;
+	union {
+		u8	byte;
+		u16	hword;
+		u32	word;
+		u64	dword;
+	} tmp;
+
+	switch (len) {
+	case 1:
+		tmp.byte	= data;
+		datap		= &tmp.byte;
+		break;
+	case 2:
+		tmp.hword	= data;
+		datap		= &tmp.hword;
+		break;
+	case 4:
+		tmp.word	= data;
+		datap		= &tmp.word;
+		break;
+	case 8:
+		tmp.dword	= data;
+		datap		= &tmp.dword;
+		break;
+	}
+
+	memcpy(buf, datap, len);
+}
+
+unsigned long kvm_mmio_read_buf(const void *buf, unsigned int len)
+{
+	unsigned long data = 0;
+	union {
+		u16	hword;
+		u32	word;
+		u64	dword;
+	} tmp;
+
+	switch (len) {
+	case 1:
+		data = *(u8 *)buf;
+		break;
+	case 2:
+		memcpy(&tmp.hword, buf, len);
+		data = tmp.hword;
+		break;
+	case 4:
+		memcpy(&tmp.word, buf, len);
+		data = tmp.word;
+		break;
+	case 8:
+		memcpy(&tmp.dword, buf, len);
+		data = tmp.dword;
+		break;
+	}
+
+	return data;
+}
+
+/**
+ * kvm_handle_mmio_return -- Handle MMIO loads after user space emulation
+ *			     or in-kernel IO emulation
+ *
+ * @vcpu: The VCPU pointer
+ * @run:  The VCPU run struct containing the mmio data
+ */
+int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
+{
+	unsigned long data;
+	unsigned int len;
+	int mask;
+
+	/* Detect an already handled MMIO return */
+	if (unlikely(!vcpu->mmio_needed))
+		return 0;
+
+	vcpu->mmio_needed = 0;
+
+	if (!run->mmio.is_write) {
+		len = run->mmio.len;
+		if (len > sizeof(unsigned long))
+			return -EINVAL;
+
+		data = kvm_mmio_read_buf(run->mmio.data, len);
+
+		if (vcpu->arch.mmio_decode.sign_extend &&
+		    len < sizeof(unsigned long)) {
+			mask = 1U << ((len * 8) - 1);
+			data = (data ^ mask) - mask;
+		}
+
+		if (!vcpu->arch.mmio_decode.sixty_four)
+			data = data & 0xffffffff;
+
+		trace_kvm_mmio(KVM_TRACE_MMIO_READ, len, run->mmio.phys_addr,
+			       &data);
+		data = vcpu_data_host_to_guest(vcpu, data, len);
+		vcpu_set_reg(vcpu, vcpu->arch.mmio_decode.rt, data);
+	}
+
+	/*
+	 * The MMIO instruction is emulated and should not be re-executed
+	 * in the guest.
+	 */
+	kvm_skip_instr(vcpu, kvm_vcpu_trap_il_is32bit(vcpu));
+
+	return 0;
+}
+
+static int decode_hsr(struct kvm_vcpu *vcpu, bool *is_write, int *len)
+{
+	unsigned long rt;
+	int access_size;
+	bool sign_extend;
+	bool sixty_four;
+
+	if (kvm_vcpu_dabt_iss1tw(vcpu)) {
+		/* page table accesses IO mem: tell guest to fix its TTBR */
+		kvm_inject_dabt(vcpu, kvm_vcpu_get_hfar(vcpu));
+		return 1;
+	}
+
+	access_size = kvm_vcpu_dabt_get_as(vcpu);
+	if (unlikely(access_size < 0))
+		return access_size;
+
+	*is_write = kvm_vcpu_dabt_iswrite(vcpu);
+	sign_extend = kvm_vcpu_dabt_issext(vcpu);
+	sixty_four = kvm_vcpu_dabt_issf(vcpu);
+	rt = kvm_vcpu_dabt_get_rd(vcpu);
+
+	*len = access_size;
+	vcpu->arch.mmio_decode.sign_extend = sign_extend;
+	vcpu->arch.mmio_decode.rt = rt;
+	vcpu->arch.mmio_decode.sixty_four = sixty_four;
+
+	return 0;
+}
+
+int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
+		 phys_addr_t fault_ipa)
+{
+	unsigned long data;
+	unsigned long rt;
+	int ret;
+	bool is_write;
+	int len;
+	u8 data_buf[8];
+
+	/*
+	 * Prepare MMIO operation. First decode the syndrome data we get
+	 * from the CPU. Then try if some in-kernel emulation feels
+	 * responsible, otherwise let user space do its magic.
+	 */
+	if (kvm_vcpu_dabt_isvalid(vcpu)) {
+		ret = decode_hsr(vcpu, &is_write, &len);
+		if (ret)
+			return ret;
+	} else {
+		kvm_err("load/store instruction decoding not implemented\n");
+		return -ENOSYS;
+	}
+
+	rt = vcpu->arch.mmio_decode.rt;
+
+	if (is_write) {
+		data = vcpu_data_guest_to_host(vcpu, vcpu_get_reg(vcpu, rt),
+					       len);
+
+		trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, len, fault_ipa, &data);
+		kvm_mmio_write_buf(data_buf, len, data);
+
+		ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+				       data_buf);
+	} else {
+		trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, len,
+			       fault_ipa, NULL);
+
+		ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+				      data_buf);
+	}
+
+	/* Now prepare kvm_run for the potential return to userland. */
+	run->mmio.is_write	= is_write;
+	run->mmio.phys_addr	= fault_ipa;
+	run->mmio.len		= len;
+	vcpu->mmio_needed	= 1;
+
+	if (!ret) {
+		/* We handled the access successfully in the kernel. */
+		if (!is_write)
+			memcpy(run->mmio.data, data_buf, len);
+		vcpu->stat.mmio_exit_kernel++;
+		kvm_handle_mmio_return(vcpu, run);
+		return 1;
+	}
+
+	if (is_write)
+		memcpy(run->mmio.data, data_buf, len);
+	vcpu->stat.mmio_exit_user++;
+	run->exit_reason	= KVM_EXIT_MMIO;
+	return 0;
+}

virt/kvm/arm/perf.c

@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Based on the x86 implementation.
+ *
+ * Copyright (C) 2012 ARM Ltd.
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ */
+
+#include <linux/perf_event.h>
+#include <linux/kvm_host.h>
+
+#include <asm/kvm_emulate.h>
+
+static int kvm_is_in_guest(void)
+{
+        return kvm_arm_get_running_vcpu() != NULL;
+}
+
+static int kvm_is_user_mode(void)
+{
+	struct kvm_vcpu *vcpu;
+
+	vcpu = kvm_arm_get_running_vcpu();
+
+	if (vcpu)
+		return !vcpu_mode_priv(vcpu);
+
+	return 0;
+}
+
+static unsigned long kvm_get_guest_ip(void)
+{
+	struct kvm_vcpu *vcpu;
+
+	vcpu = kvm_arm_get_running_vcpu();
+
+	if (vcpu)
+		return *vcpu_pc(vcpu);
+
+	return 0;
+}
+
+static struct perf_guest_info_callbacks kvm_guest_cbs = {
+	.is_in_guest	= kvm_is_in_guest,
+	.is_user_mode	= kvm_is_user_mode,
+	.get_guest_ip	= kvm_get_guest_ip,
+};
+
+int kvm_perf_init(void)
+{
+	return perf_register_guest_info_callbacks(&kvm_guest_cbs);
+}
+
+int kvm_perf_teardown(void)
+{
+	return perf_unregister_guest_info_callbacks(&kvm_guest_cbs);
+}

virt/kvm/arm/pmu.c

@@ -0,0 +1,865 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015 Linaro Ltd.
+ * Author: Shannon Zhao <shannon.zhao@linaro.org>
+ */
+
+#include <linux/cpu.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <linux/perf_event.h>
+#include <linux/perf/arm_pmu.h>
+#include <linux/uaccess.h>
+#include <asm/kvm_emulate.h>
+#include <kvm/arm_pmu.h>
+#include <kvm/arm_vgic.h>
+
+static void kvm_pmu_create_perf_event(struct kvm_vcpu *vcpu, u64 select_idx);
+
+#define PERF_ATTR_CFG1_KVM_PMU_CHAINED 0x1
+
+/**
+ * kvm_pmu_idx_is_64bit - determine if select_idx is a 64bit counter
+ * @vcpu: The vcpu pointer
+ * @select_idx: The counter index
+ */
+static bool kvm_pmu_idx_is_64bit(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	return (select_idx == ARMV8_PMU_CYCLE_IDX &&
+		__vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMU_PMCR_LC);
+}
+
+static struct kvm_vcpu *kvm_pmc_to_vcpu(struct kvm_pmc *pmc)
+{
+	struct kvm_pmu *pmu;
+	struct kvm_vcpu_arch *vcpu_arch;
+
+	pmc -= pmc->idx;
+	pmu = container_of(pmc, struct kvm_pmu, pmc[0]);
+	vcpu_arch = container_of(pmu, struct kvm_vcpu_arch, pmu);
+	return container_of(vcpu_arch, struct kvm_vcpu, arch);
+}
+
+/**
+ * kvm_pmu_pmc_is_chained - determine if the pmc is chained
+ * @pmc: The PMU counter pointer
+ */
+static bool kvm_pmu_pmc_is_chained(struct kvm_pmc *pmc)
+{
+	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
+
+	return test_bit(pmc->idx >> 1, vcpu->arch.pmu.chained);
+}
+
+/**
+ * kvm_pmu_idx_is_high_counter - determine if select_idx is a high/low counter
+ * @select_idx: The counter index
+ */
+static bool kvm_pmu_idx_is_high_counter(u64 select_idx)
+{
+	return select_idx & 0x1;
+}
+
+/**
+ * kvm_pmu_get_canonical_pmc - obtain the canonical pmc
+ * @pmc: The PMU counter pointer
+ *
+ * When a pair of PMCs are chained together we use the low counter (canonical)
+ * to hold the underlying perf event.
+ */
+static struct kvm_pmc *kvm_pmu_get_canonical_pmc(struct kvm_pmc *pmc)
+{
+	if (kvm_pmu_pmc_is_chained(pmc) &&
+	    kvm_pmu_idx_is_high_counter(pmc->idx))
+		return pmc - 1;
+
+	return pmc;
+}
+
+/**
+ * kvm_pmu_idx_has_chain_evtype - determine if the event type is chain
+ * @vcpu: The vcpu pointer
+ * @select_idx: The counter index
+ */
+static bool kvm_pmu_idx_has_chain_evtype(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	u64 eventsel, reg;
+
+	select_idx |= 0x1;
+
+	if (select_idx == ARMV8_PMU_CYCLE_IDX)
+		return false;
+
+	reg = PMEVTYPER0_EL0 + select_idx;
+	eventsel = __vcpu_sys_reg(vcpu, reg) & ARMV8_PMU_EVTYPE_EVENT;
+
+	return eventsel == ARMV8_PMUV3_PERFCTR_CHAIN;
+}
+
+/**
+ * kvm_pmu_get_pair_counter_value - get PMU counter value
+ * @vcpu: The vcpu pointer
+ * @pmc: The PMU counter pointer
+ */
+static u64 kvm_pmu_get_pair_counter_value(struct kvm_vcpu *vcpu,
+					  struct kvm_pmc *pmc)
+{
+	u64 counter, counter_high, reg, enabled, running;
+
+	if (kvm_pmu_pmc_is_chained(pmc)) {
+		pmc = kvm_pmu_get_canonical_pmc(pmc);
+		reg = PMEVCNTR0_EL0 + pmc->idx;
+
+		counter = __vcpu_sys_reg(vcpu, reg);
+		counter_high = __vcpu_sys_reg(vcpu, reg + 1);
+
+		counter = lower_32_bits(counter) | (counter_high << 32);
+	} else {
+		reg = (pmc->idx == ARMV8_PMU_CYCLE_IDX)
+		      ? PMCCNTR_EL0 : PMEVCNTR0_EL0 + pmc->idx;
+		counter = __vcpu_sys_reg(vcpu, reg);
+	}
+
+	/*
+	 * The real counter value is equal to the value of counter register plus
+	 * the value perf event counts.
+	 */
+	if (pmc->perf_event)
+		counter += perf_event_read_value(pmc->perf_event, &enabled,
+						 &running);
+
+	return counter;
+}
+
+/**
+ * kvm_pmu_get_counter_value - get PMU counter value
+ * @vcpu: The vcpu pointer
+ * @select_idx: The counter index
+ */
+u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	u64 counter;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_pmc *pmc = &pmu->pmc[select_idx];
+
+	counter = kvm_pmu_get_pair_counter_value(vcpu, pmc);
+
+	if (kvm_pmu_pmc_is_chained(pmc) &&
+	    kvm_pmu_idx_is_high_counter(select_idx))
+		counter = upper_32_bits(counter);
+	else if (select_idx != ARMV8_PMU_CYCLE_IDX)
+		counter = lower_32_bits(counter);
+
+	return counter;
+}
+
+/**
+ * kvm_pmu_set_counter_value - set PMU counter value
+ * @vcpu: The vcpu pointer
+ * @select_idx: The counter index
+ * @val: The counter value
+ */
+void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, u64 select_idx, u64 val)
+{
+	u64 reg;
+
+	reg = (select_idx == ARMV8_PMU_CYCLE_IDX)
+	      ? PMCCNTR_EL0 : PMEVCNTR0_EL0 + select_idx;
+	__vcpu_sys_reg(vcpu, reg) += (s64)val - kvm_pmu_get_counter_value(vcpu, select_idx);
+
+	/* Recreate the perf event to reflect the updated sample_period */
+	kvm_pmu_create_perf_event(vcpu, select_idx);
+}
+
+/**
+ * kvm_pmu_release_perf_event - remove the perf event
+ * @pmc: The PMU counter pointer
+ */
+static void kvm_pmu_release_perf_event(struct kvm_pmc *pmc)
+{
+	pmc = kvm_pmu_get_canonical_pmc(pmc);
+	if (pmc->perf_event) {
+		perf_event_disable(pmc->perf_event);
+		perf_event_release_kernel(pmc->perf_event);
+		pmc->perf_event = NULL;
+	}
+}
+
+/**
+ * kvm_pmu_stop_counter - stop PMU counter
+ * @pmc: The PMU counter pointer
+ *
+ * If this counter has been configured to monitor some event, release it here.
+ */
+static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, struct kvm_pmc *pmc)
+{
+	u64 counter, reg, val;
+
+	pmc = kvm_pmu_get_canonical_pmc(pmc);
+	if (!pmc->perf_event)
+		return;
+
+	counter = kvm_pmu_get_pair_counter_value(vcpu, pmc);
+
+	if (pmc->idx == ARMV8_PMU_CYCLE_IDX) {
+		reg = PMCCNTR_EL0;
+		val = counter;
+	} else {
+		reg = PMEVCNTR0_EL0 + pmc->idx;
+		val = lower_32_bits(counter);
+	}
+
+	__vcpu_sys_reg(vcpu, reg) = val;
+
+	if (kvm_pmu_pmc_is_chained(pmc))
+		__vcpu_sys_reg(vcpu, reg + 1) = upper_32_bits(counter);
+
+	kvm_pmu_release_perf_event(pmc);
+}
+
+/**
+ * kvm_pmu_vcpu_init - assign pmu counter idx for cpu
+ * @vcpu: The vcpu pointer
+ *
+ */
+void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++)
+		pmu->pmc[i].idx = i;
+}
+
+/**
+ * kvm_pmu_vcpu_reset - reset pmu state for cpu
+ * @vcpu: The vcpu pointer
+ *
+ */
+void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++)
+		kvm_pmu_stop_counter(vcpu, &pmu->pmc[i]);
+
+	bitmap_zero(vcpu->arch.pmu.chained, ARMV8_PMU_MAX_COUNTER_PAIRS);
+}
+
+/**
+ * kvm_pmu_vcpu_destroy - free perf event of PMU for cpu
+ * @vcpu: The vcpu pointer
+ *
+ */
+void kvm_pmu_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++)
+		kvm_pmu_release_perf_event(&pmu->pmc[i]);
+}
+
+u64 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu)
+{
+	u64 val = __vcpu_sys_reg(vcpu, PMCR_EL0) >> ARMV8_PMU_PMCR_N_SHIFT;
+
+	val &= ARMV8_PMU_PMCR_N_MASK;
+	if (val == 0)
+		return BIT(ARMV8_PMU_CYCLE_IDX);
+	else
+		return GENMASK(val - 1, 0) | BIT(ARMV8_PMU_CYCLE_IDX);
+}
+
+/**
+ * kvm_pmu_enable_counter_mask - enable selected PMU counters
+ * @vcpu: The vcpu pointer
+ * @val: the value guest writes to PMCNTENSET register
+ *
+ * Call perf_event_enable to start counting the perf event
+ */
+void kvm_pmu_enable_counter_mask(struct kvm_vcpu *vcpu, u64 val)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_pmc *pmc;
+
+	if (!(__vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMU_PMCR_E) || !val)
+		return;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++) {
+		if (!(val & BIT(i)))
+			continue;
+
+		pmc = &pmu->pmc[i];
+
+		/*
+		 * For high counters of chained events we must recreate the
+		 * perf event with the long (64bit) attribute set.
+		 */
+		if (kvm_pmu_pmc_is_chained(pmc) &&
+		    kvm_pmu_idx_is_high_counter(i)) {
+			kvm_pmu_create_perf_event(vcpu, i);
+			continue;
+		}
+
+		/* At this point, pmc must be the canonical */
+		if (pmc->perf_event) {
+			perf_event_enable(pmc->perf_event);
+			if (pmc->perf_event->state != PERF_EVENT_STATE_ACTIVE)
+				kvm_debug("fail to enable perf event\n");
+		}
+	}
+}
+
+/**
+ * kvm_pmu_disable_counter_mask - disable selected PMU counters
+ * @vcpu: The vcpu pointer
+ * @val: the value guest writes to PMCNTENCLR register
+ *
+ * Call perf_event_disable to stop counting the perf event
+ */
+void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_pmc *pmc;
+
+	if (!val)
+		return;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++) {
+		if (!(val & BIT(i)))
+			continue;
+
+		pmc = &pmu->pmc[i];
+
+		/*
+		 * For high counters of chained events we must recreate the
+		 * perf event with the long (64bit) attribute unset.
+		 */
+		if (kvm_pmu_pmc_is_chained(pmc) &&
+		    kvm_pmu_idx_is_high_counter(i)) {
+			kvm_pmu_create_perf_event(vcpu, i);
+			continue;
+		}
+
+		/* At this point, pmc must be the canonical */
+		if (pmc->perf_event)
+			perf_event_disable(pmc->perf_event);
+	}
+}
+
+static u64 kvm_pmu_overflow_status(struct kvm_vcpu *vcpu)
+{
+	u64 reg = 0;
+
+	if ((__vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMU_PMCR_E)) {
+		reg = __vcpu_sys_reg(vcpu, PMOVSSET_EL0);
+		reg &= __vcpu_sys_reg(vcpu, PMCNTENSET_EL0);
+		reg &= __vcpu_sys_reg(vcpu, PMINTENSET_EL1);
+		reg &= kvm_pmu_valid_counter_mask(vcpu);
+	}
+
+	return reg;
+}
+
+static void kvm_pmu_update_state(struct kvm_vcpu *vcpu)
+{
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	bool overflow;
+
+	if (!kvm_arm_pmu_v3_ready(vcpu))
+		return;
+
+	overflow = !!kvm_pmu_overflow_status(vcpu);
+	if (pmu->irq_level == overflow)
+		return;
+
+	pmu->irq_level = overflow;
+
+	if (likely(irqchip_in_kernel(vcpu->kvm))) {
+		int ret = kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id,
+					      pmu->irq_num, overflow, pmu);
+		WARN_ON(ret);
+	}
+}
+
+bool kvm_pmu_should_notify_user(struct kvm_vcpu *vcpu)
+{
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_sync_regs *sregs = &vcpu->run->s.regs;
+	bool run_level = sregs->device_irq_level & KVM_ARM_DEV_PMU;
+
+	if (likely(irqchip_in_kernel(vcpu->kvm)))
+		return false;
+
+	return pmu->irq_level != run_level;
+}
+
+/*
+ * Reflect the PMU overflow interrupt output level into the kvm_run structure
+ */
+void kvm_pmu_update_run(struct kvm_vcpu *vcpu)
+{
+	struct kvm_sync_regs *regs = &vcpu->run->s.regs;
+
+	/* Populate the timer bitmap for user space */
+	regs->device_irq_level &= ~KVM_ARM_DEV_PMU;
+	if (vcpu->arch.pmu.irq_level)
+		regs->device_irq_level |= KVM_ARM_DEV_PMU;
+}
+
+/**
+ * kvm_pmu_flush_hwstate - flush pmu state to cpu
+ * @vcpu: The vcpu pointer
+ *
+ * Check if the PMU has overflowed while we were running in the host, and inject
+ * an interrupt if that was the case.
+ */
+void kvm_pmu_flush_hwstate(struct kvm_vcpu *vcpu)
+{
+	kvm_pmu_update_state(vcpu);
+}
+
+/**
+ * kvm_pmu_sync_hwstate - sync pmu state from cpu
+ * @vcpu: The vcpu pointer
+ *
+ * Check if the PMU has overflowed while we were running in the guest, and
+ * inject an interrupt if that was the case.
+ */
+void kvm_pmu_sync_hwstate(struct kvm_vcpu *vcpu)
+{
+	kvm_pmu_update_state(vcpu);
+}
+
+/**
+ * When the perf event overflows, set the overflow status and inform the vcpu.
+ */
+static void kvm_pmu_perf_overflow(struct perf_event *perf_event,
+				  struct perf_sample_data *data,
+				  struct pt_regs *regs)
+{
+	struct kvm_pmc *pmc = perf_event->overflow_handler_context;
+	struct arm_pmu *cpu_pmu = to_arm_pmu(perf_event->pmu);
+	struct kvm_vcpu *vcpu = kvm_pmc_to_vcpu(pmc);
+	int idx = pmc->idx;
+	u64 period;
+
+	cpu_pmu->pmu.stop(perf_event, PERF_EF_UPDATE);
+
+	/*
+	 * Reset the sample period to the architectural limit,
+	 * i.e. the point where the counter overflows.
+	 */
+	period = -(local64_read(&perf_event->count));
+
+	if (!kvm_pmu_idx_is_64bit(vcpu, pmc->idx))
+		period &= GENMASK(31, 0);
+
+	local64_set(&perf_event->hw.period_left, 0);
+	perf_event->attr.sample_period = period;
+	perf_event->hw.sample_period = period;
+
+	__vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(idx);
+
+	if (kvm_pmu_overflow_status(vcpu)) {
+		kvm_make_request(KVM_REQ_IRQ_PENDING, vcpu);
+		kvm_vcpu_kick(vcpu);
+	}
+
+	cpu_pmu->pmu.start(perf_event, PERF_EF_RELOAD);
+}
+
+/**
+ * kvm_pmu_software_increment - do software increment
+ * @vcpu: The vcpu pointer
+ * @val: the value guest writes to PMSWINC register
+ */
+void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u64 val)
+{
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	int i;
+
+	if (!(__vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMU_PMCR_E))
+		return;
+
+	/* Weed out disabled counters */
+	val &= __vcpu_sys_reg(vcpu, PMCNTENSET_EL0);
+
+	for (i = 0; i < ARMV8_PMU_CYCLE_IDX; i++) {
+		u64 type, reg;
+
+		if (!(val & BIT(i)))
+			continue;
+
+		/* PMSWINC only applies to ... SW_INC! */
+		type = __vcpu_sys_reg(vcpu, PMEVTYPER0_EL0 + i);
+		type &= ARMV8_PMU_EVTYPE_EVENT;
+		if (type != ARMV8_PMUV3_PERFCTR_SW_INCR)
+			continue;
+
+		/* increment this even SW_INC counter */
+		reg = __vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i) + 1;
+		reg = lower_32_bits(reg);
+		__vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i) = reg;
+
+		if (reg) /* no overflow on the low part */
+			continue;
+
+		if (kvm_pmu_pmc_is_chained(&pmu->pmc[i])) {
+			/* increment the high counter */
+			reg = __vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i + 1) + 1;
+			reg = lower_32_bits(reg);
+			__vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i + 1) = reg;
+			if (!reg) /* mark overflow on the high counter */
+				__vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(i + 1);
+		} else {
+			/* mark overflow on low counter */
+			__vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(i);
+		}
+	}
+}
+
+/**
+ * kvm_pmu_handle_pmcr - handle PMCR register
+ * @vcpu: The vcpu pointer
+ * @val: the value guest writes to PMCR register
+ */
+void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val)
+{
+	u64 mask;
+	int i;
+
+	mask = kvm_pmu_valid_counter_mask(vcpu);
+	if (val & ARMV8_PMU_PMCR_E) {
+		kvm_pmu_enable_counter_mask(vcpu,
+		       __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) & mask);
+	} else {
+		kvm_pmu_disable_counter_mask(vcpu, mask);
+	}
+
+	if (val & ARMV8_PMU_PMCR_C)
+		kvm_pmu_set_counter_value(vcpu, ARMV8_PMU_CYCLE_IDX, 0);
+
+	if (val & ARMV8_PMU_PMCR_P) {
+		for (i = 0; i < ARMV8_PMU_CYCLE_IDX; i++)
+			kvm_pmu_set_counter_value(vcpu, i, 0);
+	}
+}
+
+static bool kvm_pmu_counter_is_enabled(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	return (__vcpu_sys_reg(vcpu, PMCR_EL0) & ARMV8_PMU_PMCR_E) &&
+	       (__vcpu_sys_reg(vcpu, PMCNTENSET_EL0) & BIT(select_idx));
+}
+
+/**
+ * kvm_pmu_create_perf_event - create a perf event for a counter
+ * @vcpu: The vcpu pointer
+ * @select_idx: The number of selected counter
+ */
+static void kvm_pmu_create_perf_event(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_pmc *pmc;
+	struct perf_event *event;
+	struct perf_event_attr attr;
+	u64 eventsel, counter, reg, data;
+
+	/*
+	 * For chained counters the event type and filtering attributes are
+	 * obtained from the low/even counter. We also use this counter to
+	 * determine if the event is enabled/disabled.
+	 */
+	pmc = kvm_pmu_get_canonical_pmc(&pmu->pmc[select_idx]);
+
+	reg = (pmc->idx == ARMV8_PMU_CYCLE_IDX)
+	      ? PMCCFILTR_EL0 : PMEVTYPER0_EL0 + pmc->idx;
+	data = __vcpu_sys_reg(vcpu, reg);
+
+	kvm_pmu_stop_counter(vcpu, pmc);
+	eventsel = data & ARMV8_PMU_EVTYPE_EVENT;
+
+	/* Software increment event does't need to be backed by a perf event */
+	if (eventsel == ARMV8_PMUV3_PERFCTR_SW_INCR &&
+	    pmc->idx != ARMV8_PMU_CYCLE_IDX)
+		return;
+
+	memset(&attr, 0, sizeof(struct perf_event_attr));
+	attr.type = PERF_TYPE_RAW;
+	attr.size = sizeof(attr);
+	attr.pinned = 1;
+	attr.disabled = !kvm_pmu_counter_is_enabled(vcpu, pmc->idx);
+	attr.exclude_user = data & ARMV8_PMU_EXCLUDE_EL0 ? 1 : 0;
+	attr.exclude_kernel = data & ARMV8_PMU_EXCLUDE_EL1 ? 1 : 0;
+	attr.exclude_hv = 1; /* Don't count EL2 events */
+	attr.exclude_host = 1; /* Don't count host events */
+	attr.config = (pmc->idx == ARMV8_PMU_CYCLE_IDX) ?
+		ARMV8_PMUV3_PERFCTR_CPU_CYCLES : eventsel;
+
+	counter = kvm_pmu_get_pair_counter_value(vcpu, pmc);
+
+	if (kvm_pmu_idx_has_chain_evtype(vcpu, pmc->idx)) {
+		/**
+		 * The initial sample period (overflow count) of an event. For
+		 * chained counters we only support overflow interrupts on the
+		 * high counter.
+		 */
+		attr.sample_period = (-counter) & GENMASK(63, 0);
+		if (kvm_pmu_counter_is_enabled(vcpu, pmc->idx + 1))
+			attr.config1 |= PERF_ATTR_CFG1_KVM_PMU_CHAINED;
+
+		event = perf_event_create_kernel_counter(&attr, -1, current,
+							 kvm_pmu_perf_overflow,
+							 pmc + 1);
+	} else {
+		/* The initial sample period (overflow count) of an event. */
+		if (kvm_pmu_idx_is_64bit(vcpu, pmc->idx))
+			attr.sample_period = (-counter) & GENMASK(63, 0);
+		else
+			attr.sample_period = (-counter) & GENMASK(31, 0);
+
+		event = perf_event_create_kernel_counter(&attr, -1, current,
+						 kvm_pmu_perf_overflow, pmc);
+	}
+
+	if (IS_ERR(event)) {
+		pr_err_once("kvm: pmu event creation failed %ld\n",
+			    PTR_ERR(event));
+		return;
+	}
+
+	pmc->perf_event = event;
+}
+
+/**
+ * kvm_pmu_update_pmc_chained - update chained bitmap
+ * @vcpu: The vcpu pointer
+ * @select_idx: The number of selected counter
+ *
+ * Update the chained bitmap based on the event type written in the
+ * typer register.
+ */
+static void kvm_pmu_update_pmc_chained(struct kvm_vcpu *vcpu, u64 select_idx)
+{
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+	struct kvm_pmc *pmc = &pmu->pmc[select_idx];
+
+	if (kvm_pmu_idx_has_chain_evtype(vcpu, pmc->idx)) {
+		/*
+		 * During promotion from !chained to chained we must ensure
+		 * the adjacent counter is stopped and its event destroyed
+		 */
+		if (!kvm_pmu_pmc_is_chained(pmc))
+			kvm_pmu_stop_counter(vcpu, pmc);
+
+		set_bit(pmc->idx >> 1, vcpu->arch.pmu.chained);
+	} else {
+		clear_bit(pmc->idx >> 1, vcpu->arch.pmu.chained);
+	}
+}
+
+/**
+ * kvm_pmu_set_counter_event_type - set selected counter to monitor some event
+ * @vcpu: The vcpu pointer
+ * @data: The data guest writes to PMXEVTYPER_EL0
+ * @select_idx: The number of selected counter
+ *
+ * When OS accesses PMXEVTYPER_EL0, that means it wants to set a PMC to count an
+ * event with given hardware event number. Here we call perf_event API to
+ * emulate this action and create a kernel perf event for it.
+ */
+void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
+				    u64 select_idx)
+{
+	u64 reg, event_type = data & ARMV8_PMU_EVTYPE_MASK;
+
+	reg = (select_idx == ARMV8_PMU_CYCLE_IDX)
+	      ? PMCCFILTR_EL0 : PMEVTYPER0_EL0 + select_idx;
+
+	__vcpu_sys_reg(vcpu, reg) = event_type;
+
+	kvm_pmu_update_pmc_chained(vcpu, select_idx);
+	kvm_pmu_create_perf_event(vcpu, select_idx);
+}
+
+bool kvm_arm_support_pmu_v3(void)
+{
+	/*
+	 * Check if HW_PERF_EVENTS are supported by checking the number of
+	 * hardware performance counters. This could ensure the presence of
+	 * a physical PMU and CONFIG_PERF_EVENT is selected.
+	 */
+	return (perf_num_counters() > 0);
+}
+
+int kvm_arm_pmu_v3_enable(struct kvm_vcpu *vcpu)
+{
+	if (!vcpu->arch.pmu.created)
+		return 0;
+
+	/*
+	 * A valid interrupt configuration for the PMU is either to have a
+	 * properly configured interrupt number and using an in-kernel
+	 * irqchip, or to not have an in-kernel GIC and not set an IRQ.
+	 */
+	if (irqchip_in_kernel(vcpu->kvm)) {
+		int irq = vcpu->arch.pmu.irq_num;
+		if (!kvm_arm_pmu_irq_initialized(vcpu))
+			return -EINVAL;
+
+		/*
+		 * If we are using an in-kernel vgic, at this point we know
+		 * the vgic will be initialized, so we can check the PMU irq
+		 * number against the dimensions of the vgic and make sure
+		 * it's valid.
+		 */
+		if (!irq_is_ppi(irq) && !vgic_valid_spi(vcpu->kvm, irq))
+			return -EINVAL;
+	} else if (kvm_arm_pmu_irq_initialized(vcpu)) {
+		   return -EINVAL;
+	}
+
+	kvm_pmu_vcpu_reset(vcpu);
+	vcpu->arch.pmu.ready = true;
+
+	return 0;
+}
+
+static int kvm_arm_pmu_v3_init(struct kvm_vcpu *vcpu)
+{
+	if (!kvm_arm_support_pmu_v3())
+		return -ENODEV;
+
+	if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+		return -ENXIO;
+
+	if (vcpu->arch.pmu.created)
+		return -EBUSY;
+
+	if (irqchip_in_kernel(vcpu->kvm)) {
+		int ret;
+
+		/*
+		 * If using the PMU with an in-kernel virtual GIC
+		 * implementation, we require the GIC to be already
+		 * initialized when initializing the PMU.
+		 */
+		if (!vgic_initialized(vcpu->kvm))
+			return -ENODEV;
+
+		if (!kvm_arm_pmu_irq_initialized(vcpu))
+			return -ENXIO;
+
+		ret = kvm_vgic_set_owner(vcpu, vcpu->arch.pmu.irq_num,
+					 &vcpu->arch.pmu);
+		if (ret)
+			return ret;
+	}
+
+	vcpu->arch.pmu.created = true;
+	return 0;
+}
+
+/*
+ * For one VM the interrupt type must be same for each vcpu.
+ * As a PPI, the interrupt number is the same for all vcpus,
+ * while as an SPI it must be a separate number per vcpu.
+ */
+static bool pmu_irq_is_valid(struct kvm *kvm, int irq)
+{
+	int i;
+	struct kvm_vcpu *vcpu;
+
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		if (!kvm_arm_pmu_irq_initialized(vcpu))
+			continue;
+
+		if (irq_is_ppi(irq)) {
+			if (vcpu->arch.pmu.irq_num != irq)
+				return false;
+		} else {
+			if (vcpu->arch.pmu.irq_num == irq)
+				return false;
+		}
+	}
+
+	return true;
+}
+
+int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
+{
+	switch (attr->attr) {
+	case KVM_ARM_VCPU_PMU_V3_IRQ: {
+		int __user *uaddr = (int __user *)(long)attr->addr;
+		int irq;
+
+		if (!irqchip_in_kernel(vcpu->kvm))
+			return -EINVAL;
+
+		if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+			return -ENODEV;
+
+		if (get_user(irq, uaddr))
+			return -EFAULT;
+
+		/* The PMU overflow interrupt can be a PPI or a valid SPI. */
+		if (!(irq_is_ppi(irq) || irq_is_spi(irq)))
+			return -EINVAL;
+
+		if (!pmu_irq_is_valid(vcpu->kvm, irq))
+			return -EINVAL;
+
+		if (kvm_arm_pmu_irq_initialized(vcpu))
+			return -EBUSY;
+
+		kvm_debug("Set kvm ARM PMU irq: %d\n", irq);
+		vcpu->arch.pmu.irq_num = irq;
+		return 0;
+	}
+	case KVM_ARM_VCPU_PMU_V3_INIT:
+		return kvm_arm_pmu_v3_init(vcpu);
+	}
+
+	return -ENXIO;
+}
+
+int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
+{
+	switch (attr->attr) {
+	case KVM_ARM_VCPU_PMU_V3_IRQ: {
+		int __user *uaddr = (int __user *)(long)attr->addr;
+		int irq;
+
+		if (!irqchip_in_kernel(vcpu->kvm))
+			return -EINVAL;
+
+		if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+			return -ENODEV;
+
+		if (!kvm_arm_pmu_irq_initialized(vcpu))
+			return -ENXIO;
+
+		irq = vcpu->arch.pmu.irq_num;
+		return put_user(irq, uaddr);
+	}
+	}
+
+	return -ENXIO;
+}
+
+int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
+{
+	switch (attr->attr) {
+	case KVM_ARM_VCPU_PMU_V3_IRQ:
+	case KVM_ARM_VCPU_PMU_V3_INIT:
+		if (kvm_arm_support_pmu_v3() &&
+		    test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+			return 0;
+	}
+
+	return -ENXIO;
+}

virt/kvm/arm/psci.c

@@ -0,0 +1,605 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2012 - ARM Ltd
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ */
+
+#include <linux/arm-smccc.h>
+#include <linux/preempt.h>
+#include <linux/kvm_host.h>
+#include <linux/uaccess.h>
+#include <linux/wait.h>
+
+#include <asm/cputype.h>
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_host.h>
+
+#include <kvm/arm_psci.h>
+
+/*
+ * This is an implementation of the Power State Coordination Interface
+ * as described in ARM document number ARM DEN 0022A.
+ */
+
+#define AFFINITY_MASK(level)	~((0x1UL << ((level) * MPIDR_LEVEL_BITS)) - 1)
+
+static u32 smccc_get_function(struct kvm_vcpu *vcpu)
+{
+	return vcpu_get_reg(vcpu, 0);
+}
+
+static unsigned long smccc_get_arg1(struct kvm_vcpu *vcpu)
+{
+	return vcpu_get_reg(vcpu, 1);
+}
+
+static unsigned long smccc_get_arg2(struct kvm_vcpu *vcpu)
+{
+	return vcpu_get_reg(vcpu, 2);
+}
+
+static unsigned long smccc_get_arg3(struct kvm_vcpu *vcpu)
+{
+	return vcpu_get_reg(vcpu, 3);
+}
+
+static void smccc_set_retval(struct kvm_vcpu *vcpu,
+			     unsigned long a0,
+			     unsigned long a1,
+			     unsigned long a2,
+			     unsigned long a3)
+{
+	vcpu_set_reg(vcpu, 0, a0);
+	vcpu_set_reg(vcpu, 1, a1);
+	vcpu_set_reg(vcpu, 2, a2);
+	vcpu_set_reg(vcpu, 3, a3);
+}
+
+static unsigned long psci_affinity_mask(unsigned long affinity_level)
+{
+	if (affinity_level <= 3)
+		return MPIDR_HWID_BITMASK & AFFINITY_MASK(affinity_level);
+
+	return 0;
+}
+
+static unsigned long kvm_psci_vcpu_suspend(struct kvm_vcpu *vcpu)
+{
+	/*
+	 * NOTE: For simplicity, we make VCPU suspend emulation to be
+	 * same-as WFI (Wait-for-interrupt) emulation.
+	 *
+	 * This means for KVM the wakeup events are interrupts and
+	 * this is consistent with intended use of StateID as described
+	 * in section 5.4.1 of PSCI v0.2 specification (ARM DEN 0022A).
+	 *
+	 * Further, we also treat power-down request to be same as
+	 * stand-by request as-per section 5.4.2 clause 3 of PSCI v0.2
+	 * specification (ARM DEN 0022A). This means all suspend states
+	 * for KVM will preserve the register state.
+	 */
+	kvm_vcpu_block(vcpu);
+	kvm_clear_request(KVM_REQ_UNHALT, vcpu);
+
+	return PSCI_RET_SUCCESS;
+}
+
+static void kvm_psci_vcpu_off(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.power_off = true;
+	kvm_make_request(KVM_REQ_SLEEP, vcpu);
+	kvm_vcpu_kick(vcpu);
+}
+
+static unsigned long kvm_psci_vcpu_on(struct kvm_vcpu *source_vcpu)
+{
+	struct vcpu_reset_state *reset_state;
+	struct kvm *kvm = source_vcpu->kvm;
+	struct kvm_vcpu *vcpu = NULL;
+	unsigned long cpu_id;
+
+	cpu_id = smccc_get_arg1(source_vcpu) & MPIDR_HWID_BITMASK;
+	if (vcpu_mode_is_32bit(source_vcpu))
+		cpu_id &= ~((u32) 0);
+
+	vcpu = kvm_mpidr_to_vcpu(kvm, cpu_id);
+
+	/*
+	 * Make sure the caller requested a valid CPU and that the CPU is
+	 * turned off.
+	 */
+	if (!vcpu)
+		return PSCI_RET_INVALID_PARAMS;
+	if (!vcpu->arch.power_off) {
+		if (kvm_psci_version(source_vcpu, kvm) != KVM_ARM_PSCI_0_1)
+			return PSCI_RET_ALREADY_ON;
+		else
+			return PSCI_RET_INVALID_PARAMS;
+	}
+
+	reset_state = &vcpu->arch.reset_state;
+
+	reset_state->pc = smccc_get_arg2(source_vcpu);
+
+	/* Propagate caller endianness */
+	reset_state->be = kvm_vcpu_is_be(source_vcpu);
+
+	/*
+	 * NOTE: We always update r0 (or x0) because for PSCI v0.1
+	 * the general puspose registers are undefined upon CPU_ON.
+	 */
+	reset_state->r0 = smccc_get_arg3(source_vcpu);
+
+	WRITE_ONCE(reset_state->reset, true);
+	kvm_make_request(KVM_REQ_VCPU_RESET, vcpu);
+
+	/*
+	 * Make sure the reset request is observed if the change to
+	 * power_state is observed.
+	 */
+	smp_wmb();
+
+	vcpu->arch.power_off = false;
+	kvm_vcpu_wake_up(vcpu);
+
+	return PSCI_RET_SUCCESS;
+}
+
+static unsigned long kvm_psci_vcpu_affinity_info(struct kvm_vcpu *vcpu)
+{
+	int i, matching_cpus = 0;
+	unsigned long mpidr;
+	unsigned long target_affinity;
+	unsigned long target_affinity_mask;
+	unsigned long lowest_affinity_level;
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_vcpu *tmp;
+
+	target_affinity = smccc_get_arg1(vcpu);
+	lowest_affinity_level = smccc_get_arg2(vcpu);
+
+	/* Determine target affinity mask */
+	target_affinity_mask = psci_affinity_mask(lowest_affinity_level);
+	if (!target_affinity_mask)
+		return PSCI_RET_INVALID_PARAMS;
+
+	/* Ignore other bits of target affinity */
+	target_affinity &= target_affinity_mask;
+
+	/*
+	 * If one or more VCPU matching target affinity are running
+	 * then ON else OFF
+	 */
+	kvm_for_each_vcpu(i, tmp, kvm) {
+		mpidr = kvm_vcpu_get_mpidr_aff(tmp);
+		if ((mpidr & target_affinity_mask) == target_affinity) {
+			matching_cpus++;
+			if (!tmp->arch.power_off)
+				return PSCI_0_2_AFFINITY_LEVEL_ON;
+		}
+	}
+
+	if (!matching_cpus)
+		return PSCI_RET_INVALID_PARAMS;
+
+	return PSCI_0_2_AFFINITY_LEVEL_OFF;
+}
+
+static void kvm_prepare_system_event(struct kvm_vcpu *vcpu, u32 type)
+{
+	int i;
+	struct kvm_vcpu *tmp;
+
+	/*
+	 * The KVM ABI specifies that a system event exit may call KVM_RUN
+	 * again and may perform shutdown/reboot at a later time that when the
+	 * actual request is made.  Since we are implementing PSCI and a
+	 * caller of PSCI reboot and shutdown expects that the system shuts
+	 * down or reboots immediately, let's make sure that VCPUs are not run
+	 * after this call is handled and before the VCPUs have been
+	 * re-initialized.
+	 */
+	kvm_for_each_vcpu(i, tmp, vcpu->kvm)
+		tmp->arch.power_off = true;
+	kvm_make_all_cpus_request(vcpu->kvm, KVM_REQ_SLEEP);
+
+	memset(&vcpu->run->system_event, 0, sizeof(vcpu->run->system_event));
+	vcpu->run->system_event.type = type;
+	vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
+}
+
+static void kvm_psci_system_off(struct kvm_vcpu *vcpu)
+{
+	kvm_prepare_system_event(vcpu, KVM_SYSTEM_EVENT_SHUTDOWN);
+}
+
+static void kvm_psci_system_reset(struct kvm_vcpu *vcpu)
+{
+	kvm_prepare_system_event(vcpu, KVM_SYSTEM_EVENT_RESET);
+}
+
+static int kvm_psci_0_2_call(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+	u32 psci_fn = smccc_get_function(vcpu);
+	unsigned long val;
+	int ret = 1;
+
+	switch (psci_fn) {
+	case PSCI_0_2_FN_PSCI_VERSION:
+		/*
+		 * Bits[31:16] = Major Version = 0
+		 * Bits[15:0] = Minor Version = 2
+		 */
+		val = KVM_ARM_PSCI_0_2;
+		break;
+	case PSCI_0_2_FN_CPU_SUSPEND:
+	case PSCI_0_2_FN64_CPU_SUSPEND:
+		val = kvm_psci_vcpu_suspend(vcpu);
+		break;
+	case PSCI_0_2_FN_CPU_OFF:
+		kvm_psci_vcpu_off(vcpu);
+		val = PSCI_RET_SUCCESS;
+		break;
+	case PSCI_0_2_FN_CPU_ON:
+	case PSCI_0_2_FN64_CPU_ON:
+		mutex_lock(&kvm->lock);
+		val = kvm_psci_vcpu_on(vcpu);
+		mutex_unlock(&kvm->lock);
+		break;
+	case PSCI_0_2_FN_AFFINITY_INFO:
+	case PSCI_0_2_FN64_AFFINITY_INFO:
+		val = kvm_psci_vcpu_affinity_info(vcpu);
+		break;
+	case PSCI_0_2_FN_MIGRATE_INFO_TYPE:
+		/*
+		 * Trusted OS is MP hence does not require migration
+	         * or
+		 * Trusted OS is not present
+		 */
+		val = PSCI_0_2_TOS_MP;
+		break;
+	case PSCI_0_2_FN_SYSTEM_OFF:
+		kvm_psci_system_off(vcpu);
+		/*
+		 * We should'nt be going back to guest VCPU after
+		 * receiving SYSTEM_OFF request.
+		 *
+		 * If user space accidently/deliberately resumes
+		 * guest VCPU after SYSTEM_OFF request then guest
+		 * VCPU should see internal failure from PSCI return
+		 * value. To achieve this, we preload r0 (or x0) with
+		 * PSCI return value INTERNAL_FAILURE.
+		 */
+		val = PSCI_RET_INTERNAL_FAILURE;
+		ret = 0;
+		break;
+	case PSCI_0_2_FN_SYSTEM_RESET:
+		kvm_psci_system_reset(vcpu);
+		/*
+		 * Same reason as SYSTEM_OFF for preloading r0 (or x0)
+		 * with PSCI return value INTERNAL_FAILURE.
+		 */
+		val = PSCI_RET_INTERNAL_FAILURE;
+		ret = 0;
+		break;
+	default:
+		val = PSCI_RET_NOT_SUPPORTED;
+		break;
+	}
+
+	smccc_set_retval(vcpu, val, 0, 0, 0);
+	return ret;
+}
+
+static int kvm_psci_1_0_call(struct kvm_vcpu *vcpu)
+{
+	u32 psci_fn = smccc_get_function(vcpu);
+	u32 feature;
+	unsigned long val;
+	int ret = 1;
+
+	switch(psci_fn) {
+	case PSCI_0_2_FN_PSCI_VERSION:
+		val = KVM_ARM_PSCI_1_0;
+		break;
+	case PSCI_1_0_FN_PSCI_FEATURES:
+		feature = smccc_get_arg1(vcpu);
+		switch(feature) {
+		case PSCI_0_2_FN_PSCI_VERSION:
+		case PSCI_0_2_FN_CPU_SUSPEND:
+		case PSCI_0_2_FN64_CPU_SUSPEND:
+		case PSCI_0_2_FN_CPU_OFF:
+		case PSCI_0_2_FN_CPU_ON:
+		case PSCI_0_2_FN64_CPU_ON:
+		case PSCI_0_2_FN_AFFINITY_INFO:
+		case PSCI_0_2_FN64_AFFINITY_INFO:
+		case PSCI_0_2_FN_MIGRATE_INFO_TYPE:
+		case PSCI_0_2_FN_SYSTEM_OFF:
+		case PSCI_0_2_FN_SYSTEM_RESET:
+		case PSCI_1_0_FN_PSCI_FEATURES:
+		case ARM_SMCCC_VERSION_FUNC_ID:
+			val = 0;
+			break;
+		default:
+			val = PSCI_RET_NOT_SUPPORTED;
+			break;
+		}
+		break;
+	default:
+		return kvm_psci_0_2_call(vcpu);
+	}
+
+	smccc_set_retval(vcpu, val, 0, 0, 0);
+	return ret;
+}
+
+static int kvm_psci_0_1_call(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+	u32 psci_fn = smccc_get_function(vcpu);
+	unsigned long val;
+
+	switch (psci_fn) {
+	case KVM_PSCI_FN_CPU_OFF:
+		kvm_psci_vcpu_off(vcpu);
+		val = PSCI_RET_SUCCESS;
+		break;
+	case KVM_PSCI_FN_CPU_ON:
+		mutex_lock(&kvm->lock);
+		val = kvm_psci_vcpu_on(vcpu);
+		mutex_unlock(&kvm->lock);
+		break;
+	default:
+		val = PSCI_RET_NOT_SUPPORTED;
+		break;
+	}
+
+	smccc_set_retval(vcpu, val, 0, 0, 0);
+	return 1;
+}
+
+/**
+ * kvm_psci_call - handle PSCI call if r0 value is in range
+ * @vcpu: Pointer to the VCPU struct
+ *
+ * Handle PSCI calls from guests through traps from HVC instructions.
+ * The calling convention is similar to SMC calls to the secure world
+ * where the function number is placed in r0.
+ *
+ * This function returns: > 0 (success), 0 (success but exit to user
+ * space), and < 0 (errors)
+ *
+ * Errors:
+ * -EINVAL: Unrecognized PSCI function
+ */
+static int kvm_psci_call(struct kvm_vcpu *vcpu)
+{
+	switch (kvm_psci_version(vcpu, vcpu->kvm)) {
+	case KVM_ARM_PSCI_1_0:
+		return kvm_psci_1_0_call(vcpu);
+	case KVM_ARM_PSCI_0_2:
+		return kvm_psci_0_2_call(vcpu);
+	case KVM_ARM_PSCI_0_1:
+		return kvm_psci_0_1_call(vcpu);
+	default:
+		return -EINVAL;
+	};
+}
+
+int kvm_hvc_call_handler(struct kvm_vcpu *vcpu)
+{
+	u32 func_id = smccc_get_function(vcpu);
+	u32 val = SMCCC_RET_NOT_SUPPORTED;
+	u32 feature;
+
+	switch (func_id) {
+	case ARM_SMCCC_VERSION_FUNC_ID:
+		val = ARM_SMCCC_VERSION_1_1;
+		break;
+	case ARM_SMCCC_ARCH_FEATURES_FUNC_ID:
+		feature = smccc_get_arg1(vcpu);
+		switch(feature) {
+		case ARM_SMCCC_ARCH_WORKAROUND_1:
+			switch (kvm_arm_harden_branch_predictor()) {
+			case KVM_BP_HARDEN_UNKNOWN:
+				break;
+			case KVM_BP_HARDEN_WA_NEEDED:
+				val = SMCCC_RET_SUCCESS;
+				break;
+			case KVM_BP_HARDEN_NOT_REQUIRED:
+				val = SMCCC_RET_NOT_REQUIRED;
+				break;
+			}
+			break;
+		case ARM_SMCCC_ARCH_WORKAROUND_2:
+			switch (kvm_arm_have_ssbd()) {
+			case KVM_SSBD_FORCE_DISABLE:
+			case KVM_SSBD_UNKNOWN:
+				break;
+			case KVM_SSBD_KERNEL:
+				val = SMCCC_RET_SUCCESS;
+				break;
+			case KVM_SSBD_FORCE_ENABLE:
+			case KVM_SSBD_MITIGATED:
+				val = SMCCC_RET_NOT_REQUIRED;
+				break;
+			}
+			break;
+		}
+		break;
+	default:
+		return kvm_psci_call(vcpu);
+	}
+
+	smccc_set_retval(vcpu, val, 0, 0, 0);
+	return 1;
+}
+
+int kvm_arm_get_fw_num_regs(struct kvm_vcpu *vcpu)
+{
+	return 3;		/* PSCI version and two workaround registers */
+}
+
+int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices)
+{
+	if (put_user(KVM_REG_ARM_PSCI_VERSION, uindices++))
+		return -EFAULT;
+
+	if (put_user(KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1, uindices++))
+		return -EFAULT;
+
+	if (put_user(KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2, uindices++))
+		return -EFAULT;
+
+	return 0;
+}
+
+#define KVM_REG_FEATURE_LEVEL_WIDTH	4
+#define KVM_REG_FEATURE_LEVEL_MASK	(BIT(KVM_REG_FEATURE_LEVEL_WIDTH) - 1)
+
+/*
+ * Convert the workaround level into an easy-to-compare number, where higher
+ * values mean better protection.
+ */
+static int get_kernel_wa_level(u64 regid)
+{
+	switch (regid) {
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+		switch (kvm_arm_harden_branch_predictor()) {
+		case KVM_BP_HARDEN_UNKNOWN:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
+		case KVM_BP_HARDEN_WA_NEEDED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_AVAIL;
+		case KVM_BP_HARDEN_NOT_REQUIRED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_REQUIRED;
+		}
+		return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1_NOT_AVAIL;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+		switch (kvm_arm_have_ssbd()) {
+		case KVM_SSBD_FORCE_DISABLE:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL;
+		case KVM_SSBD_KERNEL:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL;
+		case KVM_SSBD_FORCE_ENABLE:
+		case KVM_SSBD_MITIGATED:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED;
+		case KVM_SSBD_UNKNOWN:
+		default:
+			return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_UNKNOWN;
+		}
+	}
+
+	return -EINVAL;
+}
+
+int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
+{
+	void __user *uaddr = (void __user *)(long)reg->addr;
+	u64 val;
+
+	switch (reg->id) {
+	case KVM_REG_ARM_PSCI_VERSION:
+		val = kvm_psci_version(vcpu, vcpu->kvm);
+		break;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+		val = get_kernel_wa_level(reg->id) & KVM_REG_FEATURE_LEVEL_MASK;
+		break;
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+		val = get_kernel_wa_level(reg->id) & KVM_REG_FEATURE_LEVEL_MASK;
+
+		if (val == KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL &&
+		    kvm_arm_get_vcpu_workaround_2_flag(vcpu))
+			val |= KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED;
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	if (copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	return 0;
+}
+
+int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
+{
+	void __user *uaddr = (void __user *)(long)reg->addr;
+	u64 val;
+	int wa_level;
+
+	if (copy_from_user(&val, uaddr, KVM_REG_SIZE(reg->id)))
+		return -EFAULT;
+
+	switch (reg->id) {
+	case KVM_REG_ARM_PSCI_VERSION:
+	{
+		bool wants_02;
+
+		wants_02 = test_bit(KVM_ARM_VCPU_PSCI_0_2, vcpu->arch.features);
+
+		switch (val) {
+		case KVM_ARM_PSCI_0_1:
+			if (wants_02)
+				return -EINVAL;
+			vcpu->kvm->arch.psci_version = val;
+			return 0;
+		case KVM_ARM_PSCI_0_2:
+		case KVM_ARM_PSCI_1_0:
+			if (!wants_02)
+				return -EINVAL;
+			vcpu->kvm->arch.psci_version = val;
+			return 0;
+		}
+		break;
+	}
+
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1:
+		if (val & ~KVM_REG_FEATURE_LEVEL_MASK)
+			return -EINVAL;
+
+		if (get_kernel_wa_level(reg->id) < val)
+			return -EINVAL;
+
+		return 0;
+
+	case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2:
+		if (val & ~(KVM_REG_FEATURE_LEVEL_MASK |
+			    KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED))
+			return -EINVAL;
+
+		wa_level = val & KVM_REG_FEATURE_LEVEL_MASK;
+
+		if (get_kernel_wa_level(reg->id) < wa_level)
+			return -EINVAL;
+
+		/* The enabled bit must not be set unless the level is AVAIL. */
+		if (wa_level != KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL &&
+		    wa_level != val)
+			return -EINVAL;
+
+		/* Are we finished or do we need to check the enable bit ? */
+		if (kvm_arm_have_ssbd() != KVM_SSBD_KERNEL)
+			return 0;
+
+		/*
+		 * If this kernel supports the workaround to be switched on
+		 * or off, make sure it matches the requested setting.
+		 */
+		switch (wa_level) {
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_AVAIL:
+			kvm_arm_set_vcpu_workaround_2_flag(vcpu,
+			    val & KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_ENABLED);
+			break;
+		case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_REQUIRED:
+			kvm_arm_set_vcpu_workaround_2_flag(vcpu, true);
+			break;
+		}
+
+		return 0;
+	default:
+		return -ENOENT;
+	}
+
+	return -EINVAL;
+}

virt/kvm/arm/trace.h

@@ -0,0 +1,378 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#if !defined(_TRACE_KVM_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_KVM_H
+
+#include <kvm/arm_arch_timer.h>
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM kvm
+
+/*
+ * Tracepoints for entry/exit to guest
+ */
+TRACE_EVENT(kvm_entry,
+	TP_PROTO(unsigned long vcpu_pc),
+	TP_ARGS(vcpu_pc),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	vcpu_pc		)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc		= vcpu_pc;
+	),
+
+	TP_printk("PC: 0x%08lx", __entry->vcpu_pc)
+);
+
+TRACE_EVENT(kvm_exit,
+	TP_PROTO(int ret, unsigned int esr_ec, unsigned long vcpu_pc),
+	TP_ARGS(ret, esr_ec, vcpu_pc),
+
+	TP_STRUCT__entry(
+		__field(	int,		ret		)
+		__field(	unsigned int,	esr_ec		)
+		__field(	unsigned long,	vcpu_pc		)
+	),
+
+	TP_fast_assign(
+		__entry->ret			= ARM_EXCEPTION_CODE(ret);
+		__entry->esr_ec = ARM_EXCEPTION_IS_TRAP(ret) ? esr_ec : 0;
+		__entry->vcpu_pc		= vcpu_pc;
+	),
+
+	TP_printk("%s: HSR_EC: 0x%04x (%s), PC: 0x%08lx",
+		  __print_symbolic(__entry->ret, kvm_arm_exception_type),
+		  __entry->esr_ec,
+		  __print_symbolic(__entry->esr_ec, kvm_arm_exception_class),
+		  __entry->vcpu_pc)
+);
+
+TRACE_EVENT(kvm_guest_fault,
+	TP_PROTO(unsigned long vcpu_pc, unsigned long hsr,
+		 unsigned long hxfar,
+		 unsigned long long ipa),
+	TP_ARGS(vcpu_pc, hsr, hxfar, ipa),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	vcpu_pc		)
+		__field(	unsigned long,	hsr		)
+		__field(	unsigned long,	hxfar		)
+		__field(   unsigned long long,	ipa		)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc		= vcpu_pc;
+		__entry->hsr			= hsr;
+		__entry->hxfar			= hxfar;
+		__entry->ipa			= ipa;
+	),
+
+	TP_printk("ipa %#llx, hsr %#08lx, hxfar %#08lx, pc %#08lx",
+		  __entry->ipa, __entry->hsr,
+		  __entry->hxfar, __entry->vcpu_pc)
+);
+
+TRACE_EVENT(kvm_access_fault,
+	TP_PROTO(unsigned long ipa),
+	TP_ARGS(ipa),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	ipa		)
+	),
+
+	TP_fast_assign(
+		__entry->ipa		= ipa;
+	),
+
+	TP_printk("IPA: %lx", __entry->ipa)
+);
+
+TRACE_EVENT(kvm_irq_line,
+	TP_PROTO(unsigned int type, int vcpu_idx, int irq_num, int level),
+	TP_ARGS(type, vcpu_idx, irq_num, level),
+
+	TP_STRUCT__entry(
+		__field(	unsigned int,	type		)
+		__field(	int,		vcpu_idx	)
+		__field(	int,		irq_num		)
+		__field(	int,		level		)
+	),
+
+	TP_fast_assign(
+		__entry->type		= type;
+		__entry->vcpu_idx	= vcpu_idx;
+		__entry->irq_num	= irq_num;
+		__entry->level		= level;
+	),
+
+	TP_printk("Inject %s interrupt (%d), vcpu->idx: %d, num: %d, level: %d",
+		  (__entry->type == KVM_ARM_IRQ_TYPE_CPU) ? "CPU" :
+		  (__entry->type == KVM_ARM_IRQ_TYPE_PPI) ? "VGIC PPI" :
+		  (__entry->type == KVM_ARM_IRQ_TYPE_SPI) ? "VGIC SPI" : "UNKNOWN",
+		  __entry->type, __entry->vcpu_idx, __entry->irq_num, __entry->level)
+);
+
+TRACE_EVENT(kvm_mmio_emulate,
+	TP_PROTO(unsigned long vcpu_pc, unsigned long instr,
+		 unsigned long cpsr),
+	TP_ARGS(vcpu_pc, instr, cpsr),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	vcpu_pc		)
+		__field(	unsigned long,	instr		)
+		__field(	unsigned long,	cpsr		)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_pc		= vcpu_pc;
+		__entry->instr			= instr;
+		__entry->cpsr			= cpsr;
+	),
+
+	TP_printk("Emulate MMIO at: 0x%08lx (instr: %08lx, cpsr: %08lx)",
+		  __entry->vcpu_pc, __entry->instr, __entry->cpsr)
+);
+
+TRACE_EVENT(kvm_unmap_hva_range,
+	TP_PROTO(unsigned long start, unsigned long end),
+	TP_ARGS(start, end),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	start		)
+		__field(	unsigned long,	end		)
+	),
+
+	TP_fast_assign(
+		__entry->start		= start;
+		__entry->end		= end;
+	),
+
+	TP_printk("mmu notifier unmap range: %#08lx -- %#08lx",
+		  __entry->start, __entry->end)
+);
+
+TRACE_EVENT(kvm_set_spte_hva,
+	TP_PROTO(unsigned long hva),
+	TP_ARGS(hva),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	hva		)
+	),
+
+	TP_fast_assign(
+		__entry->hva		= hva;
+	),
+
+	TP_printk("mmu notifier set pte hva: %#08lx", __entry->hva)
+);
+
+TRACE_EVENT(kvm_age_hva,
+	TP_PROTO(unsigned long start, unsigned long end),
+	TP_ARGS(start, end),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	start		)
+		__field(	unsigned long,	end		)
+	),
+
+	TP_fast_assign(
+		__entry->start		= start;
+		__entry->end		= end;
+	),
+
+	TP_printk("mmu notifier age hva: %#08lx -- %#08lx",
+		  __entry->start, __entry->end)
+);
+
+TRACE_EVENT(kvm_test_age_hva,
+	TP_PROTO(unsigned long hva),
+	TP_ARGS(hva),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	hva		)
+	),
+
+	TP_fast_assign(
+		__entry->hva		= hva;
+	),
+
+	TP_printk("mmu notifier test age hva: %#08lx", __entry->hva)
+);
+
+TRACE_EVENT(kvm_set_way_flush,
+	    TP_PROTO(unsigned long vcpu_pc, bool cache),
+	    TP_ARGS(vcpu_pc, cache),
+
+	    TP_STRUCT__entry(
+		    __field(	unsigned long,	vcpu_pc		)
+		    __field(	bool,		cache		)
+	    ),
+
+	    TP_fast_assign(
+		    __entry->vcpu_pc		= vcpu_pc;
+		    __entry->cache		= cache;
+	    ),
+
+	    TP_printk("S/W flush at 0x%016lx (cache %s)",
+		      __entry->vcpu_pc, __entry->cache ? "on" : "off")
+);
+
+TRACE_EVENT(kvm_toggle_cache,
+	    TP_PROTO(unsigned long vcpu_pc, bool was, bool now),
+	    TP_ARGS(vcpu_pc, was, now),
+
+	    TP_STRUCT__entry(
+		    __field(	unsigned long,	vcpu_pc		)
+		    __field(	bool,		was		)
+		    __field(	bool,		now		)
+	    ),
+
+	    TP_fast_assign(
+		    __entry->vcpu_pc		= vcpu_pc;
+		    __entry->was		= was;
+		    __entry->now		= now;
+	    ),
+
+	    TP_printk("VM op at 0x%016lx (cache was %s, now %s)",
+		      __entry->vcpu_pc, __entry->was ? "on" : "off",
+		      __entry->now ? "on" : "off")
+);
+
+/*
+ * Tracepoints for arch_timer
+ */
+TRACE_EVENT(kvm_timer_update_irq,
+	TP_PROTO(unsigned long vcpu_id, __u32 irq, int level),
+	TP_ARGS(vcpu_id, irq, level),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	vcpu_id	)
+		__field(	__u32,		irq	)
+		__field(	int,		level	)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_id	= vcpu_id;
+		__entry->irq		= irq;
+		__entry->level		= level;
+	),
+
+	TP_printk("VCPU: %ld, IRQ %d, level %d",
+		  __entry->vcpu_id, __entry->irq, __entry->level)
+);
+
+TRACE_EVENT(kvm_get_timer_map,
+	TP_PROTO(unsigned long vcpu_id, struct timer_map *map),
+	TP_ARGS(vcpu_id, map),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,		vcpu_id	)
+		__field(	int,			direct_vtimer	)
+		__field(	int,			direct_ptimer	)
+		__field(	int,			emul_ptimer	)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_id		= vcpu_id;
+		__entry->direct_vtimer		= arch_timer_ctx_index(map->direct_vtimer);
+		__entry->direct_ptimer =
+			(map->direct_ptimer) ? arch_timer_ctx_index(map->direct_ptimer) : -1;
+		__entry->emul_ptimer =
+			(map->emul_ptimer) ? arch_timer_ctx_index(map->emul_ptimer) : -1;
+	),
+
+	TP_printk("VCPU: %ld, dv: %d, dp: %d, ep: %d",
+		  __entry->vcpu_id,
+		  __entry->direct_vtimer,
+		  __entry->direct_ptimer,
+		  __entry->emul_ptimer)
+);
+
+TRACE_EVENT(kvm_timer_save_state,
+	TP_PROTO(struct arch_timer_context *ctx),
+	TP_ARGS(ctx),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,		ctl		)
+		__field(	unsigned long long,	cval		)
+		__field(	int,			timer_idx	)
+	),
+
+	TP_fast_assign(
+		__entry->ctl			= ctx->cnt_ctl;
+		__entry->cval			= ctx->cnt_cval;
+		__entry->timer_idx		= arch_timer_ctx_index(ctx);
+	),
+
+	TP_printk("   CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+		  __entry->ctl,
+		  __entry->cval,
+		  __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_restore_state,
+	TP_PROTO(struct arch_timer_context *ctx),
+	TP_ARGS(ctx),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,		ctl		)
+		__field(	unsigned long long,	cval		)
+		__field(	int,			timer_idx	)
+	),
+
+	TP_fast_assign(
+		__entry->ctl			= ctx->cnt_ctl;
+		__entry->cval			= ctx->cnt_cval;
+		__entry->timer_idx		= arch_timer_ctx_index(ctx);
+	),
+
+	TP_printk("CTL: %#08lx CVAL: %#16llx arch_timer_ctx_index: %d",
+		  __entry->ctl,
+		  __entry->cval,
+		  __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_hrtimer_expire,
+	TP_PROTO(struct arch_timer_context *ctx),
+	TP_ARGS(ctx),
+
+	TP_STRUCT__entry(
+		__field(	int,			timer_idx	)
+	),
+
+	TP_fast_assign(
+		__entry->timer_idx		= arch_timer_ctx_index(ctx);
+	),
+
+	TP_printk("arch_timer_ctx_index: %d", __entry->timer_idx)
+);
+
+TRACE_EVENT(kvm_timer_emulate,
+	TP_PROTO(struct arch_timer_context *ctx, bool should_fire),
+	TP_ARGS(ctx, should_fire),
+
+	TP_STRUCT__entry(
+		__field(	int,			timer_idx	)
+		__field(	bool,			should_fire	)
+	),
+
+	TP_fast_assign(
+		__entry->timer_idx		= arch_timer_ctx_index(ctx);
+		__entry->should_fire		= should_fire;
+	),
+
+	TP_printk("arch_timer_ctx_index: %d (should_fire: %d)",
+		  __entry->timer_idx, __entry->should_fire)
+);
+
+#endif /* _TRACE_KVM_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH ../../virt/kvm/arm
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>

virt/kvm/arm/vgic/trace.h

@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#if !defined(_TRACE_VGIC_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_VGIC_H
+
+#include <linux/tracepoint.h>
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM kvm
+
+TRACE_EVENT(vgic_update_irq_pending,
+	TP_PROTO(unsigned long vcpu_id, __u32 irq, bool level),
+	TP_ARGS(vcpu_id, irq, level),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	vcpu_id	)
+		__field(	__u32,		irq	)
+		__field(	bool,		level	)
+	),
+
+	TP_fast_assign(
+		__entry->vcpu_id	= vcpu_id;
+		__entry->irq		= irq;
+		__entry->level		= level;
+	),
+
+	TP_printk("VCPU: %ld, IRQ %d, level: %d",
+		  __entry->vcpu_id, __entry->irq, __entry->level)
+);
+
+#endif /* _TRACE_VGIC_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH ../../virt/kvm/arm/vgic
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>

virt/kvm/arm/vgic/vgic-debug.c

@@ -0,0 +1,288 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2016 Linaro
+ * Author: Christoffer Dall <christoffer.dall@linaro.org>
+ */
+
+#include <linux/cpu.h>
+#include <linux/debugfs.h>
+#include <linux/interrupt.h>
+#include <linux/kvm_host.h>
+#include <linux/seq_file.h>
+#include <kvm/arm_vgic.h>
+#include <asm/kvm_mmu.h>
+#include "vgic.h"
+
+/*
+ * Structure to control looping through the entire vgic state.  We start at
+ * zero for each field and move upwards.  So, if dist_id is 0 we print the
+ * distributor info.  When dist_id is 1, we have already printed it and move
+ * on.
+ *
+ * When vcpu_id < nr_cpus we print the vcpu info until vcpu_id == nr_cpus and
+ * so on.
+ */
+struct vgic_state_iter {
+	int nr_cpus;
+	int nr_spis;
+	int nr_lpis;
+	int dist_id;
+	int vcpu_id;
+	int intid;
+	int lpi_idx;
+	u32 *lpi_array;
+};
+
+static void iter_next(struct vgic_state_iter *iter)
+{
+	if (iter->dist_id == 0) {
+		iter->dist_id++;
+		return;
+	}
+
+	iter->intid++;
+	if (iter->intid == VGIC_NR_PRIVATE_IRQS &&
+	    ++iter->vcpu_id < iter->nr_cpus)
+		iter->intid = 0;
+
+	if (iter->intid >= (iter->nr_spis + VGIC_NR_PRIVATE_IRQS)) {
+		if (iter->lpi_idx < iter->nr_lpis)
+			iter->intid = iter->lpi_array[iter->lpi_idx];
+		iter->lpi_idx++;
+	}
+}
+
+static void iter_init(struct kvm *kvm, struct vgic_state_iter *iter,
+		      loff_t pos)
+{
+	int nr_cpus = atomic_read(&kvm->online_vcpus);
+
+	memset(iter, 0, sizeof(*iter));
+
+	iter->nr_cpus = nr_cpus;
+	iter->nr_spis = kvm->arch.vgic.nr_spis;
+	if (kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+		iter->nr_lpis = vgic_copy_lpi_list(kvm, NULL, &iter->lpi_array);
+		if (iter->nr_lpis < 0)
+			iter->nr_lpis = 0;
+	}
+
+	/* Fast forward to the right position if needed */
+	while (pos--)
+		iter_next(iter);
+}
+
+static bool end_of_vgic(struct vgic_state_iter *iter)
+{
+	return iter->dist_id > 0 &&
+		iter->vcpu_id == iter->nr_cpus &&
+		iter->intid >= (iter->nr_spis + VGIC_NR_PRIVATE_IRQS) &&
+		iter->lpi_idx > iter->nr_lpis;
+}
+
+static void *vgic_debug_start(struct seq_file *s, loff_t *pos)
+{
+	struct kvm *kvm = (struct kvm *)s->private;
+	struct vgic_state_iter *iter;
+
+	mutex_lock(&kvm->lock);
+	iter = kvm->arch.vgic.iter;
+	if (iter) {
+		iter = ERR_PTR(-EBUSY);
+		goto out;
+	}
+
+	iter = kmalloc(sizeof(*iter), GFP_KERNEL);
+	if (!iter) {
+		iter = ERR_PTR(-ENOMEM);
+		goto out;
+	}
+
+	iter_init(kvm, iter, *pos);
+	kvm->arch.vgic.iter = iter;
+
+	if (end_of_vgic(iter))
+		iter = NULL;
+out:
+	mutex_unlock(&kvm->lock);
+	return iter;
+}
+
+static void *vgic_debug_next(struct seq_file *s, void *v, loff_t *pos)
+{
+	struct kvm *kvm = (struct kvm *)s->private;
+	struct vgic_state_iter *iter = kvm->arch.vgic.iter;
+
+	++*pos;
+	iter_next(iter);
+	if (end_of_vgic(iter))
+		iter = NULL;
+	return iter;
+}
+
+static void vgic_debug_stop(struct seq_file *s, void *v)
+{
+	struct kvm *kvm = (struct kvm *)s->private;
+	struct vgic_state_iter *iter;
+
+	/*
+	 * If the seq file wasn't properly opened, there's nothing to clearn
+	 * up.
+	 */
+	if (IS_ERR(v))
+		return;
+
+	mutex_lock(&kvm->lock);
+	iter = kvm->arch.vgic.iter;
+	kfree(iter->lpi_array);
+	kfree(iter);
+	kvm->arch.vgic.iter = NULL;
+	mutex_unlock(&kvm->lock);
+}
+
+static void print_dist_state(struct seq_file *s, struct vgic_dist *dist)
+{
+	bool v3 = dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3;
+
+	seq_printf(s, "Distributor\n");
+	seq_printf(s, "===========\n");
+	seq_printf(s, "vgic_model:\t%s\n", v3 ? "GICv3" : "GICv2");
+	seq_printf(s, "nr_spis:\t%d\n", dist->nr_spis);
+	if (v3)
+		seq_printf(s, "nr_lpis:\t%d\n", dist->lpi_list_count);
+	seq_printf(s, "enabled:\t%d\n", dist->enabled);
+	seq_printf(s, "\n");
+
+	seq_printf(s, "P=pending_latch, L=line_level, A=active\n");
+	seq_printf(s, "E=enabled, H=hw, C=config (level=1, edge=0)\n");
+	seq_printf(s, "G=group\n");
+}
+
+static void print_header(struct seq_file *s, struct vgic_irq *irq,
+			 struct kvm_vcpu *vcpu)
+{
+	int id = 0;
+	char *hdr = "SPI ";
+
+	if (vcpu) {
+		hdr = "VCPU";
+		id = vcpu->vcpu_id;
+	}
+
+	seq_printf(s, "\n");
+	seq_printf(s, "%s%2d TYP   ID TGT_ID PLAEHCG     HWID   TARGET SRC PRI VCPU_ID\n", hdr, id);
+	seq_printf(s, "----------------------------------------------------------------\n");
+}
+
+static void print_irq_state(struct seq_file *s, struct vgic_irq *irq,
+			    struct kvm_vcpu *vcpu)
+{
+	char *type;
+	if (irq->intid < VGIC_NR_SGIS)
+		type = "SGI";
+	else if (irq->intid < VGIC_NR_PRIVATE_IRQS)
+		type = "PPI";
+	else if (irq->intid < VGIC_MAX_SPI)
+		type = "SPI";
+	else
+		type = "LPI";
+
+	if (irq->intid ==0 || irq->intid == VGIC_NR_PRIVATE_IRQS)
+		print_header(s, irq, vcpu);
+
+	seq_printf(s, "       %s %4d "
+		      "    %2d "
+		      "%d%d%d%d%d%d%d "
+		      "%8d "
+		      "%8x "
+		      " %2x "
+		      "%3d "
+		      "     %2d "
+		      "\n",
+			type, irq->intid,
+			(irq->target_vcpu) ? irq->target_vcpu->vcpu_id : -1,
+			irq->pending_latch,
+			irq->line_level,
+			irq->active,
+			irq->enabled,
+			irq->hw,
+			irq->config == VGIC_CONFIG_LEVEL,
+			irq->group,
+			irq->hwintid,
+			irq->mpidr,
+			irq->source,
+			irq->priority,
+			(irq->vcpu) ? irq->vcpu->vcpu_id : -1);
+}
+
+static int vgic_debug_show(struct seq_file *s, void *v)
+{
+	struct kvm *kvm = (struct kvm *)s->private;
+	struct vgic_state_iter *iter = (struct vgic_state_iter *)v;
+	struct vgic_irq *irq;
+	struct kvm_vcpu *vcpu = NULL;
+	unsigned long flags;
+
+	if (iter->dist_id == 0) {
+		print_dist_state(s, &kvm->arch.vgic);
+		return 0;
+	}
+
+	if (!kvm->arch.vgic.initialized)
+		return 0;
+
+	if (iter->vcpu_id < iter->nr_cpus)
+		vcpu = kvm_get_vcpu(kvm, iter->vcpu_id);
+
+	irq = vgic_get_irq(kvm, vcpu, iter->intid);
+	if (!irq) {
+		seq_printf(s, "       LPI %4d freed\n", iter->intid);
+		return 0;
+	}
+
+	raw_spin_lock_irqsave(&irq->irq_lock, flags);
+	print_irq_state(s, irq, vcpu);
+	raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+
+	vgic_put_irq(kvm, irq);
+	return 0;
+}
+
+static const struct seq_operations vgic_debug_seq_ops = {
+	.start = vgic_debug_start,
+	.next  = vgic_debug_next,
+	.stop  = vgic_debug_stop,
+	.show  = vgic_debug_show
+};
+
+static int debug_open(struct inode *inode, struct file *file)
+{
+	int ret;
+	ret = seq_open(file, &vgic_debug_seq_ops);
+	if (!ret) {
+		struct seq_file *seq;
+		/* seq_open will have modified file->private_data */
+		seq = file->private_data;
+		seq->private = inode->i_private;
+	}
+
+	return ret;
+};
+
+static const struct file_operations vgic_debug_fops = {
+	.owner   = THIS_MODULE,
+	.open    = debug_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = seq_release
+};
+
+void vgic_debug_init(struct kvm *kvm)
+{
+	debugfs_create_file("vgic-state", 0444, kvm->debugfs_dentry, kvm,
+			    &vgic_debug_fops);
+}
+
+void vgic_debug_destroy(struct kvm *kvm)
+{
+}

virt/kvm/arm/vgic/vgic-init.c

@@ -0,0 +1,552 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ */
+
+#include <linux/uaccess.h>
+#include <linux/interrupt.h>
+#include <linux/cpu.h>
+#include <linux/kvm_host.h>
+#include <kvm/arm_vgic.h>
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_mmu.h>
+#include "vgic.h"
+
+/*
+ * Initialization rules: there are multiple stages to the vgic
+ * initialization, both for the distributor and the CPU interfaces.  The basic
+ * idea is that even though the VGIC is not functional or not requested from
+ * user space, the critical path of the run loop can still call VGIC functions
+ * that just won't do anything, without them having to check additional
+ * initialization flags to ensure they don't look at uninitialized data
+ * structures.
+ *
+ * Distributor:
+ *
+ * - kvm_vgic_early_init(): initialization of static data that doesn't
+ *   depend on any sizing information or emulation type. No allocation
+ *   is allowed there.
+ *
+ * - vgic_init(): allocation and initialization of the generic data
+ *   structures that depend on sizing information (number of CPUs,
+ *   number of interrupts). Also initializes the vcpu specific data
+ *   structures. Can be executed lazily for GICv2.
+ *
+ * CPU Interface:
+ *
+ * - kvm_vgic_vcpu_init(): initialization of static data that
+ *   doesn't depend on any sizing information or emulation type. No
+ *   allocation is allowed there.
+ */
+
+/* EARLY INIT */
+
+/**
+ * kvm_vgic_early_init() - Initialize static VGIC VCPU data structures
+ * @kvm: The VM whose VGIC districutor should be initialized
+ *
+ * Only do initialization of static structures that don't require any
+ * allocation or sizing information from userspace.  vgic_init() called
+ * kvm_vgic_dist_init() which takes care of the rest.
+ */
+void kvm_vgic_early_init(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+
+	INIT_LIST_HEAD(&dist->lpi_list_head);
+	INIT_LIST_HEAD(&dist->lpi_translation_cache);
+	raw_spin_lock_init(&dist->lpi_list_lock);
+}
+
+/* CREATION */
+
+/**
+ * kvm_vgic_create: triggered by the instantiation of the VGIC device by
+ * user space, either through the legacy KVM_CREATE_IRQCHIP ioctl (v2 only)
+ * or through the generic KVM_CREATE_DEVICE API ioctl.
+ * irqchip_in_kernel() tells you if this function succeeded or not.
+ * @kvm: kvm struct pointer
+ * @type: KVM_DEV_TYPE_ARM_VGIC_V[23]
+ */
+int kvm_vgic_create(struct kvm *kvm, u32 type)
+{
+	int i, vcpu_lock_idx = -1, ret;
+	struct kvm_vcpu *vcpu;
+
+	if (irqchip_in_kernel(kvm))
+		return -EEXIST;
+
+	/*
+	 * This function is also called by the KVM_CREATE_IRQCHIP handler,
+	 * which had no chance yet to check the availability of the GICv2
+	 * emulation. So check this here again. KVM_CREATE_DEVICE does
+	 * the proper checks already.
+	 */
+	if (type == KVM_DEV_TYPE_ARM_VGIC_V2 &&
+		!kvm_vgic_global_state.can_emulate_gicv2)
+		return -ENODEV;
+
+	/*
+	 * Any time a vcpu is run, vcpu_load is called which tries to grab the
+	 * vcpu->mutex.  By grabbing the vcpu->mutex of all VCPUs we ensure
+	 * that no other VCPUs are run while we create the vgic.
+	 */
+	ret = -EBUSY;
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		if (!mutex_trylock(&vcpu->mutex))
+			goto out_unlock;
+		vcpu_lock_idx = i;
+	}
+
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		if (vcpu->arch.has_run_once)
+			goto out_unlock;
+	}
+	ret = 0;
+
+	if (type == KVM_DEV_TYPE_ARM_VGIC_V2)
+		kvm->arch.max_vcpus = VGIC_V2_MAX_CPUS;
+	else
+		kvm->arch.max_vcpus = VGIC_V3_MAX_CPUS;
+
+	if (atomic_read(&kvm->online_vcpus) > kvm->arch.max_vcpus) {
+		ret = -E2BIG;
+		goto out_unlock;
+	}
+
+	kvm->arch.vgic.in_kernel = true;
+	kvm->arch.vgic.vgic_model = type;
+
+	kvm->arch.vgic.vgic_dist_base = VGIC_ADDR_UNDEF;
+
+	if (type == KVM_DEV_TYPE_ARM_VGIC_V2)
+		kvm->arch.vgic.vgic_cpu_base = VGIC_ADDR_UNDEF;
+	else
+		INIT_LIST_HEAD(&kvm->arch.vgic.rd_regions);
+
+out_unlock:
+	for (; vcpu_lock_idx >= 0; vcpu_lock_idx--) {
+		vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx);
+		mutex_unlock(&vcpu->mutex);
+	}
+	return ret;
+}
+
+/* INIT/DESTROY */
+
+/**
+ * kvm_vgic_dist_init: initialize the dist data structures
+ * @kvm: kvm struct pointer
+ * @nr_spis: number of spis, frozen by caller
+ */
+static int kvm_vgic_dist_init(struct kvm *kvm, unsigned int nr_spis)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu0 = kvm_get_vcpu(kvm, 0);
+	int i;
+
+	dist->spis = kcalloc(nr_spis, sizeof(struct vgic_irq), GFP_KERNEL);
+	if (!dist->spis)
+		return  -ENOMEM;
+
+	/*
+	 * In the following code we do not take the irq struct lock since
+	 * no other action on irq structs can happen while the VGIC is
+	 * not initialized yet:
+	 * If someone wants to inject an interrupt or does a MMIO access, we
+	 * require prior initialization in case of a virtual GICv3 or trigger
+	 * initialization when using a virtual GICv2.
+	 */
+	for (i = 0; i < nr_spis; i++) {
+		struct vgic_irq *irq = &dist->spis[i];
+
+		irq->intid = i + VGIC_NR_PRIVATE_IRQS;
+		INIT_LIST_HEAD(&irq->ap_list);
+		raw_spin_lock_init(&irq->irq_lock);
+		irq->vcpu = NULL;
+		irq->target_vcpu = vcpu0;
+		kref_init(&irq->refcount);
+		switch (dist->vgic_model) {
+		case KVM_DEV_TYPE_ARM_VGIC_V2:
+			irq->targets = 0;
+			irq->group = 0;
+			break;
+		case KVM_DEV_TYPE_ARM_VGIC_V3:
+			irq->mpidr = 0;
+			irq->group = 1;
+			break;
+		default:
+			kfree(dist->spis);
+			return -EINVAL;
+		}
+	}
+	return 0;
+}
+
+/**
+ * kvm_vgic_vcpu_init() - Initialize static VGIC VCPU data
+ * structures and register VCPU-specific KVM iodevs
+ *
+ * @vcpu: pointer to the VCPU being created and initialized
+ *
+ * Only do initialization, but do not actually enable the
+ * VGIC CPU interface
+ */
+int kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	int ret = 0;
+	int i;
+
+	vgic_cpu->rd_iodev.base_addr = VGIC_ADDR_UNDEF;
+
+	INIT_LIST_HEAD(&vgic_cpu->ap_list_head);
+	raw_spin_lock_init(&vgic_cpu->ap_list_lock);
+
+	/*
+	 * Enable and configure all SGIs to be edge-triggered and
+	 * configure all PPIs as level-triggered.
+	 */
+	for (i = 0; i < VGIC_NR_PRIVATE_IRQS; i++) {
+		struct vgic_irq *irq = &vgic_cpu->private_irqs[i];
+
+		INIT_LIST_HEAD(&irq->ap_list);
+		raw_spin_lock_init(&irq->irq_lock);
+		irq->intid = i;
+		irq->vcpu = NULL;
+		irq->target_vcpu = vcpu;
+		kref_init(&irq->refcount);
+		if (vgic_irq_is_sgi(i)) {
+			/* SGIs */
+			irq->enabled = 1;
+			irq->config = VGIC_CONFIG_EDGE;
+		} else {
+			/* PPIs */
+			irq->config = VGIC_CONFIG_LEVEL;
+		}
+	}
+
+	if (!irqchip_in_kernel(vcpu->kvm))
+		return 0;
+
+	/*
+	 * If we are creating a VCPU with a GICv3 we must also register the
+	 * KVM io device for the redistributor that belongs to this VCPU.
+	 */
+	if (dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+		mutex_lock(&vcpu->kvm->lock);
+		ret = vgic_register_redist_iodev(vcpu);
+		mutex_unlock(&vcpu->kvm->lock);
+	}
+	return ret;
+}
+
+static void kvm_vgic_vcpu_enable(struct kvm_vcpu *vcpu)
+{
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_enable(vcpu);
+	else
+		vgic_v3_enable(vcpu);
+}
+
+/*
+ * vgic_init: allocates and initializes dist and vcpu data structures
+ * depending on two dimensioning parameters:
+ * - the number of spis
+ * - the number of vcpus
+ * The function is generally called when nr_spis has been explicitly set
+ * by the guest through the KVM DEVICE API. If not nr_spis is set to 256.
+ * vgic_initialized() returns true when this function has succeeded.
+ * Must be called with kvm->lock held!
+ */
+int vgic_init(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu;
+	int ret = 0, i, idx;
+
+	if (vgic_initialized(kvm))
+		return 0;
+
+	/* Are we also in the middle of creating a VCPU? */
+	if (kvm->created_vcpus != atomic_read(&kvm->online_vcpus))
+		return -EBUSY;
+
+	/* freeze the number of spis */
+	if (!dist->nr_spis)
+		dist->nr_spis = VGIC_NR_IRQS_LEGACY - VGIC_NR_PRIVATE_IRQS;
+
+	ret = kvm_vgic_dist_init(kvm, dist->nr_spis);
+	if (ret)
+		goto out;
+
+	/* Initialize groups on CPUs created before the VGIC type was known */
+	kvm_for_each_vcpu(idx, vcpu, kvm) {
+		struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+		for (i = 0; i < VGIC_NR_PRIVATE_IRQS; i++) {
+			struct vgic_irq *irq = &vgic_cpu->private_irqs[i];
+			switch (dist->vgic_model) {
+			case KVM_DEV_TYPE_ARM_VGIC_V3:
+				irq->group = 1;
+				irq->mpidr = kvm_vcpu_get_mpidr_aff(vcpu);
+				break;
+			case KVM_DEV_TYPE_ARM_VGIC_V2:
+				irq->group = 0;
+				irq->targets = 1U << idx;
+				break;
+			default:
+				ret = -EINVAL;
+				goto out;
+			}
+		}
+	}
+
+	if (vgic_has_its(kvm)) {
+		vgic_lpi_translation_cache_init(kvm);
+		ret = vgic_v4_init(kvm);
+		if (ret)
+			goto out;
+	}
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vgic_vcpu_enable(vcpu);
+
+	ret = kvm_vgic_setup_default_irq_routing(kvm);
+	if (ret)
+		goto out;
+
+	vgic_debug_init(kvm);
+
+	dist->implementation_rev = 2;
+	dist->initialized = true;
+
+out:
+	return ret;
+}
+
+static void kvm_vgic_dist_destroy(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct vgic_redist_region *rdreg, *next;
+
+	dist->ready = false;
+	dist->initialized = false;
+
+	kfree(dist->spis);
+	dist->spis = NULL;
+	dist->nr_spis = 0;
+
+	if (kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+		list_for_each_entry_safe(rdreg, next, &dist->rd_regions, list) {
+			list_del(&rdreg->list);
+			kfree(rdreg);
+		}
+		INIT_LIST_HEAD(&dist->rd_regions);
+	}
+
+	if (vgic_has_its(kvm))
+		vgic_lpi_translation_cache_destroy(kvm);
+
+	if (vgic_supports_direct_msis(kvm))
+		vgic_v4_teardown(kvm);
+}
+
+void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	INIT_LIST_HEAD(&vgic_cpu->ap_list_head);
+}
+
+/* To be called with kvm->lock held */
+static void __kvm_vgic_destroy(struct kvm *kvm)
+{
+	struct kvm_vcpu *vcpu;
+	int i;
+
+	vgic_debug_destroy(kvm);
+
+	kvm_vgic_dist_destroy(kvm);
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_vgic_vcpu_destroy(vcpu);
+}
+
+void kvm_vgic_destroy(struct kvm *kvm)
+{
+	mutex_lock(&kvm->lock);
+	__kvm_vgic_destroy(kvm);
+	mutex_unlock(&kvm->lock);
+}
+
+/**
+ * vgic_lazy_init: Lazy init is only allowed if the GIC exposed to the guest
+ * is a GICv2. A GICv3 must be explicitly initialized by the guest using the
+ * KVM_DEV_ARM_VGIC_GRP_CTRL KVM_DEVICE group.
+ * @kvm: kvm struct pointer
+ */
+int vgic_lazy_init(struct kvm *kvm)
+{
+	int ret = 0;
+
+	if (unlikely(!vgic_initialized(kvm))) {
+		/*
+		 * We only provide the automatic initialization of the VGIC
+		 * for the legacy case of a GICv2. Any other type must
+		 * be explicitly initialized once setup with the respective
+		 * KVM device call.
+		 */
+		if (kvm->arch.vgic.vgic_model != KVM_DEV_TYPE_ARM_VGIC_V2)
+			return -EBUSY;
+
+		mutex_lock(&kvm->lock);
+		ret = vgic_init(kvm);
+		mutex_unlock(&kvm->lock);
+	}
+
+	return ret;
+}
+
+/* RESOURCE MAPPING */
+
+/**
+ * Map the MMIO regions depending on the VGIC model exposed to the guest
+ * called on the first VCPU run.
+ * Also map the virtual CPU interface into the VM.
+ * v2/v3 derivatives call vgic_init if not already done.
+ * vgic_ready() returns true if this function has succeeded.
+ * @kvm: kvm struct pointer
+ */
+int kvm_vgic_map_resources(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	int ret = 0;
+
+	mutex_lock(&kvm->lock);
+	if (!irqchip_in_kernel(kvm))
+		goto out;
+
+	if (dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V2)
+		ret = vgic_v2_map_resources(kvm);
+	else
+		ret = vgic_v3_map_resources(kvm);
+
+	if (ret)
+		__kvm_vgic_destroy(kvm);
+
+out:
+	mutex_unlock(&kvm->lock);
+	return ret;
+}
+
+/* GENERIC PROBE */
+
+static int vgic_init_cpu_starting(unsigned int cpu)
+{
+	enable_percpu_irq(kvm_vgic_global_state.maint_irq, 0);
+	return 0;
+}
+
+
+static int vgic_init_cpu_dying(unsigned int cpu)
+{
+	disable_percpu_irq(kvm_vgic_global_state.maint_irq);
+	return 0;
+}
+
+static irqreturn_t vgic_maintenance_handler(int irq, void *data)
+{
+	/*
+	 * We cannot rely on the vgic maintenance interrupt to be
+	 * delivered synchronously. This means we can only use it to
+	 * exit the VM, and we perform the handling of EOIed
+	 * interrupts on the exit path (see vgic_fold_lr_state).
+	 */
+	return IRQ_HANDLED;
+}
+
+/**
+ * kvm_vgic_init_cpu_hardware - initialize the GIC VE hardware
+ *
+ * For a specific CPU, initialize the GIC VE hardware.
+ */
+void kvm_vgic_init_cpu_hardware(void)
+{
+	BUG_ON(preemptible());
+
+	/*
+	 * We want to make sure the list registers start out clear so that we
+	 * only have the program the used registers.
+	 */
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_init_lrs();
+	else
+		kvm_call_hyp(__vgic_v3_init_lrs);
+}
+
+/**
+ * kvm_vgic_hyp_init: populates the kvm_vgic_global_state variable
+ * according to the host GIC model. Accordingly calls either
+ * vgic_v2/v3_probe which registers the KVM_DEVICE that can be
+ * instantiated by a guest later on .
+ */
+int kvm_vgic_hyp_init(void)
+{
+	const struct gic_kvm_info *gic_kvm_info;
+	int ret;
+
+	gic_kvm_info = gic_get_kvm_info();
+	if (!gic_kvm_info)
+		return -ENODEV;
+
+	if (!gic_kvm_info->maint_irq) {
+		kvm_err("No vgic maintenance irq\n");
+		return -ENXIO;
+	}
+
+	switch (gic_kvm_info->type) {
+	case GIC_V2:
+		ret = vgic_v2_probe(gic_kvm_info);
+		break;
+	case GIC_V3:
+		ret = vgic_v3_probe(gic_kvm_info);
+		if (!ret) {
+			static_branch_enable(&kvm_vgic_global_state.gicv3_cpuif);
+			kvm_info("GIC system register CPU interface enabled\n");
+		}
+		break;
+	default:
+		ret = -ENODEV;
+	}
+
+	if (ret)
+		return ret;
+
+	kvm_vgic_global_state.maint_irq = gic_kvm_info->maint_irq;
+	ret = request_percpu_irq(kvm_vgic_global_state.maint_irq,
+				 vgic_maintenance_handler,
+				 "vgic", kvm_get_running_vcpus());
+	if (ret) {
+		kvm_err("Cannot register interrupt %d\n",
+			kvm_vgic_global_state.maint_irq);
+		return ret;
+	}
+
+	ret = cpuhp_setup_state(CPUHP_AP_KVM_ARM_VGIC_INIT_STARTING,
+				"kvm/arm/vgic:starting",
+				vgic_init_cpu_starting, vgic_init_cpu_dying);
+	if (ret) {
+		kvm_err("Cannot register vgic CPU notifier\n");
+		goto out_free_irq;
+	}
+
+	kvm_info("vgic interrupt IRQ%d\n", kvm_vgic_global_state.maint_irq);
+	return 0;
+
+out_free_irq:
+	free_percpu_irq(kvm_vgic_global_state.maint_irq,
+			kvm_get_running_vcpus());
+	return ret;
+}

virt/kvm/arm/vgic/vgic-irqfd.c

@@ -0,0 +1,141 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ */
+
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <trace/events/kvm.h>
+#include <kvm/arm_vgic.h>
+#include "vgic.h"
+
+/**
+ * vgic_irqfd_set_irq: inject the IRQ corresponding to the
+ * irqchip routing entry
+ *
+ * This is the entry point for irqfd IRQ injection
+ */
+static int vgic_irqfd_set_irq(struct kvm_kernel_irq_routing_entry *e,
+			struct kvm *kvm, int irq_source_id,
+			int level, bool line_status)
+{
+	unsigned int spi_id = e->irqchip.pin + VGIC_NR_PRIVATE_IRQS;
+
+	if (!vgic_valid_spi(kvm, spi_id))
+		return -EINVAL;
+	return kvm_vgic_inject_irq(kvm, 0, spi_id, level, NULL);
+}
+
+/**
+ * kvm_set_routing_entry: populate a kvm routing entry
+ * from a user routing entry
+ *
+ * @kvm: the VM this entry is applied to
+ * @e: kvm kernel routing entry handle
+ * @ue: user api routing entry handle
+ * return 0 on success, -EINVAL on errors.
+ */
+int kvm_set_routing_entry(struct kvm *kvm,
+			  struct kvm_kernel_irq_routing_entry *e,
+			  const struct kvm_irq_routing_entry *ue)
+{
+	int r = -EINVAL;
+
+	switch (ue->type) {
+	case KVM_IRQ_ROUTING_IRQCHIP:
+		e->set = vgic_irqfd_set_irq;
+		e->irqchip.irqchip = ue->u.irqchip.irqchip;
+		e->irqchip.pin = ue->u.irqchip.pin;
+		if ((e->irqchip.pin >= KVM_IRQCHIP_NUM_PINS) ||
+		    (e->irqchip.irqchip >= KVM_NR_IRQCHIPS))
+			goto out;
+		break;
+	case KVM_IRQ_ROUTING_MSI:
+		e->set = kvm_set_msi;
+		e->msi.address_lo = ue->u.msi.address_lo;
+		e->msi.address_hi = ue->u.msi.address_hi;
+		e->msi.data = ue->u.msi.data;
+		e->msi.flags = ue->flags;
+		e->msi.devid = ue->u.msi.devid;
+		break;
+	default:
+		goto out;
+	}
+	r = 0;
+out:
+	return r;
+}
+
+static void kvm_populate_msi(struct kvm_kernel_irq_routing_entry *e,
+			     struct kvm_msi *msi)
+{
+	msi->address_lo = e->msi.address_lo;
+	msi->address_hi = e->msi.address_hi;
+	msi->data = e->msi.data;
+	msi->flags = e->msi.flags;
+	msi->devid = e->msi.devid;
+}
+/**
+ * kvm_set_msi: inject the MSI corresponding to the
+ * MSI routing entry
+ *
+ * This is the entry point for irqfd MSI injection
+ * and userspace MSI injection.
+ */
+int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e,
+		struct kvm *kvm, int irq_source_id,
+		int level, bool line_status)
+{
+	struct kvm_msi msi;
+
+	if (!vgic_has_its(kvm))
+		return -ENODEV;
+
+	if (!level)
+		return -1;
+
+	kvm_populate_msi(e, &msi);
+	return vgic_its_inject_msi(kvm, &msi);
+}
+
+/**
+ * kvm_arch_set_irq_inatomic: fast-path for irqfd injection
+ *
+ * Currently only direct MSI injection is supported.
+ */
+int kvm_arch_set_irq_inatomic(struct kvm_kernel_irq_routing_entry *e,
+			      struct kvm *kvm, int irq_source_id, int level,
+			      bool line_status)
+{
+	if (e->type == KVM_IRQ_ROUTING_MSI && vgic_has_its(kvm) && level) {
+		struct kvm_msi msi;
+
+		kvm_populate_msi(e, &msi);
+		if (!vgic_its_inject_cached_translation(kvm, &msi))
+			return 0;
+	}
+
+	return -EWOULDBLOCK;
+}
+
+int kvm_vgic_setup_default_irq_routing(struct kvm *kvm)
+{
+	struct kvm_irq_routing_entry *entries;
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	u32 nr = dist->nr_spis;
+	int i, ret;
+
+	entries = kcalloc(nr, sizeof(*entries), GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	for (i = 0; i < nr; i++) {
+		entries[i].gsi = i;
+		entries[i].type = KVM_IRQ_ROUTING_IRQCHIP;
+		entries[i].u.irqchip.irqchip = 0;
+		entries[i].u.irqchip.pin = i;
+	}
+	ret = kvm_set_irq_routing(kvm, entries, nr, 0);
+	kfree(entries);
+	return ret;
+}

virt/kvm/arm/vgic/vgic-kvm-device.c

@@ -0,0 +1,741 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * VGIC: KVM DEVICE API
+ *
+ * Copyright (C) 2015 ARM Ltd.
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ */
+#include <linux/kvm_host.h>
+#include <kvm/arm_vgic.h>
+#include <linux/uaccess.h>
+#include <asm/kvm_mmu.h>
+#include <asm/cputype.h>
+#include "vgic.h"
+
+/* common helpers */
+
+int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
+		      phys_addr_t addr, phys_addr_t alignment)
+{
+	if (addr & ~kvm_phys_mask(kvm))
+		return -E2BIG;
+
+	if (!IS_ALIGNED(addr, alignment))
+		return -EINVAL;
+
+	if (!IS_VGIC_ADDR_UNDEF(*ioaddr))
+		return -EEXIST;
+
+	return 0;
+}
+
+static int vgic_check_type(struct kvm *kvm, int type_needed)
+{
+	if (kvm->arch.vgic.vgic_model != type_needed)
+		return -ENODEV;
+	else
+		return 0;
+}
+
+/**
+ * kvm_vgic_addr - set or get vgic VM base addresses
+ * @kvm:   pointer to the vm struct
+ * @type:  the VGIC addr type, one of KVM_VGIC_V[23]_ADDR_TYPE_XXX
+ * @addr:  pointer to address value
+ * @write: if true set the address in the VM address space, if false read the
+ *          address
+ *
+ * Set or get the vgic base addresses for the distributor and the virtual CPU
+ * interface in the VM physical address space.  These addresses are properties
+ * of the emulated core/SoC and therefore user space initially knows this
+ * information.
+ * Check them for sanity (alignment, double assignment). We can't check for
+ * overlapping regions in case of a virtual GICv3 here, since we don't know
+ * the number of VCPUs yet, so we defer this check to map_resources().
+ */
+int kvm_vgic_addr(struct kvm *kvm, unsigned long type, u64 *addr, bool write)
+{
+	int r = 0;
+	struct vgic_dist *vgic = &kvm->arch.vgic;
+	phys_addr_t *addr_ptr, alignment;
+	u64 undef_value = VGIC_ADDR_UNDEF;
+
+	mutex_lock(&kvm->lock);
+	switch (type) {
+	case KVM_VGIC_V2_ADDR_TYPE_DIST:
+		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V2);
+		addr_ptr = &vgic->vgic_dist_base;
+		alignment = SZ_4K;
+		break;
+	case KVM_VGIC_V2_ADDR_TYPE_CPU:
+		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V2);
+		addr_ptr = &vgic->vgic_cpu_base;
+		alignment = SZ_4K;
+		break;
+	case KVM_VGIC_V3_ADDR_TYPE_DIST:
+		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V3);
+		addr_ptr = &vgic->vgic_dist_base;
+		alignment = SZ_64K;
+		break;
+	case KVM_VGIC_V3_ADDR_TYPE_REDIST: {
+		struct vgic_redist_region *rdreg;
+
+		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V3);
+		if (r)
+			break;
+		if (write) {
+			r = vgic_v3_set_redist_base(kvm, 0, *addr, 0);
+			goto out;
+		}
+		rdreg = list_first_entry(&vgic->rd_regions,
+					 struct vgic_redist_region, list);
+		if (!rdreg)
+			addr_ptr = &undef_value;
+		else
+			addr_ptr = &rdreg->base;
+		break;
+	}
+	case KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION:
+	{
+		struct vgic_redist_region *rdreg;
+		u8 index;
+
+		r = vgic_check_type(kvm, KVM_DEV_TYPE_ARM_VGIC_V3);
+		if (r)
+			break;
+
+		index = *addr & KVM_VGIC_V3_RDIST_INDEX_MASK;
+
+		if (write) {
+			gpa_t base = *addr & KVM_VGIC_V3_RDIST_BASE_MASK;
+			u32 count = (*addr & KVM_VGIC_V3_RDIST_COUNT_MASK)
+					>> KVM_VGIC_V3_RDIST_COUNT_SHIFT;
+			u8 flags = (*addr & KVM_VGIC_V3_RDIST_FLAGS_MASK)
+					>> KVM_VGIC_V3_RDIST_FLAGS_SHIFT;
+
+			if (!count || flags)
+				r = -EINVAL;
+			else
+				r = vgic_v3_set_redist_base(kvm, index,
+							    base, count);
+			goto out;
+		}
+
+		rdreg = vgic_v3_rdist_region_from_index(kvm, index);
+		if (!rdreg) {
+			r = -ENOENT;
+			goto out;
+		}
+
+		*addr = index;
+		*addr |= rdreg->base;
+		*addr |= (u64)rdreg->count << KVM_VGIC_V3_RDIST_COUNT_SHIFT;
+		goto out;
+	}
+	default:
+		r = -ENODEV;
+	}
+
+	if (r)
+		goto out;
+
+	if (write) {
+		r = vgic_check_ioaddr(kvm, addr_ptr, *addr, alignment);
+		if (!r)
+			*addr_ptr = *addr;
+	} else {
+		*addr = *addr_ptr;
+	}
+
+out:
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+
+static int vgic_set_common_attr(struct kvm_device *dev,
+				struct kvm_device_attr *attr)
+{
+	int r;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR: {
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		u64 addr;
+		unsigned long type = (unsigned long)attr->attr;
+
+		if (copy_from_user(&addr, uaddr, sizeof(addr)))
+			return -EFAULT;
+
+		r = kvm_vgic_addr(dev->kvm, type, &addr, true);
+		return (r == -ENODEV) ? -ENXIO : r;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u32 val;
+		int ret = 0;
+
+		if (get_user(val, uaddr))
+			return -EFAULT;
+
+		/*
+		 * We require:
+		 * - at least 32 SPIs on top of the 16 SGIs and 16 PPIs
+		 * - at most 1024 interrupts
+		 * - a multiple of 32 interrupts
+		 */
+		if (val < (VGIC_NR_PRIVATE_IRQS + 32) ||
+		    val > VGIC_MAX_RESERVED ||
+		    (val & 31))
+			return -EINVAL;
+
+		mutex_lock(&dev->kvm->lock);
+
+		if (vgic_ready(dev->kvm) || dev->kvm->arch.vgic.nr_spis)
+			ret = -EBUSY;
+		else
+			dev->kvm->arch.vgic.nr_spis =
+				val - VGIC_NR_PRIVATE_IRQS;
+
+		mutex_unlock(&dev->kvm->lock);
+
+		return ret;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CTRL: {
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_CTRL_INIT:
+			mutex_lock(&dev->kvm->lock);
+			r = vgic_init(dev->kvm);
+			mutex_unlock(&dev->kvm->lock);
+			return r;
+		}
+		break;
+	}
+	}
+
+	return -ENXIO;
+}
+
+static int vgic_get_common_attr(struct kvm_device *dev,
+				struct kvm_device_attr *attr)
+{
+	int r = -ENXIO;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR: {
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		u64 addr;
+		unsigned long type = (unsigned long)attr->attr;
+
+		r = kvm_vgic_addr(dev->kvm, type, &addr, false);
+		if (r)
+			return (r == -ENODEV) ? -ENXIO : r;
+
+		if (copy_to_user(uaddr, &addr, sizeof(addr)))
+			return -EFAULT;
+		break;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+
+		r = put_user(dev->kvm->arch.vgic.nr_spis +
+			     VGIC_NR_PRIVATE_IRQS, uaddr);
+		break;
+	}
+	}
+
+	return r;
+}
+
+static int vgic_create(struct kvm_device *dev, u32 type)
+{
+	return kvm_vgic_create(dev->kvm, type);
+}
+
+static void vgic_destroy(struct kvm_device *dev)
+{
+	kfree(dev);
+}
+
+int kvm_register_vgic_device(unsigned long type)
+{
+	int ret = -ENODEV;
+
+	switch (type) {
+	case KVM_DEV_TYPE_ARM_VGIC_V2:
+		ret = kvm_register_device_ops(&kvm_arm_vgic_v2_ops,
+					      KVM_DEV_TYPE_ARM_VGIC_V2);
+		break;
+	case KVM_DEV_TYPE_ARM_VGIC_V3:
+		ret = kvm_register_device_ops(&kvm_arm_vgic_v3_ops,
+					      KVM_DEV_TYPE_ARM_VGIC_V3);
+
+		if (ret)
+			break;
+		ret = kvm_vgic_register_its_device();
+		break;
+	}
+
+	return ret;
+}
+
+int vgic_v2_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr,
+		       struct vgic_reg_attr *reg_attr)
+{
+	int cpuid;
+
+	cpuid = (attr->attr & KVM_DEV_ARM_VGIC_CPUID_MASK) >>
+		 KVM_DEV_ARM_VGIC_CPUID_SHIFT;
+
+	if (cpuid >= atomic_read(&dev->kvm->online_vcpus))
+		return -EINVAL;
+
+	reg_attr->vcpu = kvm_get_vcpu(dev->kvm, cpuid);
+	reg_attr->addr = attr->attr & KVM_DEV_ARM_VGIC_OFFSET_MASK;
+
+	return 0;
+}
+
+/* unlocks vcpus from @vcpu_lock_idx and smaller */
+static void unlock_vcpus(struct kvm *kvm, int vcpu_lock_idx)
+{
+	struct kvm_vcpu *tmp_vcpu;
+
+	for (; vcpu_lock_idx >= 0; vcpu_lock_idx--) {
+		tmp_vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx);
+		mutex_unlock(&tmp_vcpu->mutex);
+	}
+}
+
+void unlock_all_vcpus(struct kvm *kvm)
+{
+	unlock_vcpus(kvm, atomic_read(&kvm->online_vcpus) - 1);
+}
+
+/* Returns true if all vcpus were locked, false otherwise */
+bool lock_all_vcpus(struct kvm *kvm)
+{
+	struct kvm_vcpu *tmp_vcpu;
+	int c;
+
+	/*
+	 * Any time a vcpu is run, vcpu_load is called which tries to grab the
+	 * vcpu->mutex.  By grabbing the vcpu->mutex of all VCPUs we ensure
+	 * that no other VCPUs are run and fiddle with the vgic state while we
+	 * access it.
+	 */
+	kvm_for_each_vcpu(c, tmp_vcpu, kvm) {
+		if (!mutex_trylock(&tmp_vcpu->mutex)) {
+			unlock_vcpus(kvm, c - 1);
+			return false;
+		}
+	}
+
+	return true;
+}
+
+/**
+ * vgic_v2_attr_regs_access - allows user space to access VGIC v2 state
+ *
+ * @dev:      kvm device handle
+ * @attr:     kvm device attribute
+ * @reg:      address the value is read or written
+ * @is_write: true if userspace is writing a register
+ */
+static int vgic_v2_attr_regs_access(struct kvm_device *dev,
+				    struct kvm_device_attr *attr,
+				    u32 *reg, bool is_write)
+{
+	struct vgic_reg_attr reg_attr;
+	gpa_t addr;
+	struct kvm_vcpu *vcpu;
+	int ret;
+
+	ret = vgic_v2_parse_attr(dev, attr, &reg_attr);
+	if (ret)
+		return ret;
+
+	vcpu = reg_attr.vcpu;
+	addr = reg_attr.addr;
+
+	mutex_lock(&dev->kvm->lock);
+
+	ret = vgic_init(dev->kvm);
+	if (ret)
+		goto out;
+
+	if (!lock_all_vcpus(dev->kvm)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS:
+		ret = vgic_v2_cpuif_uaccess(vcpu, is_write, addr, reg);
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+		ret = vgic_v2_dist_uaccess(vcpu, is_write, addr, reg);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	unlock_all_vcpus(dev->kvm);
+out:
+	mutex_unlock(&dev->kvm->lock);
+	return ret;
+}
+
+static int vgic_v2_set_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	int ret;
+
+	ret = vgic_set_common_attr(dev, attr);
+	if (ret != -ENXIO)
+		return ret;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u32 reg;
+
+		if (get_user(reg, uaddr))
+			return -EFAULT;
+
+		return vgic_v2_attr_regs_access(dev, attr, &reg, true);
+	}
+	}
+
+	return -ENXIO;
+}
+
+static int vgic_v2_get_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	int ret;
+
+	ret = vgic_get_common_attr(dev, attr);
+	if (ret != -ENXIO)
+		return ret;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u32 reg = 0;
+
+		ret = vgic_v2_attr_regs_access(dev, attr, &reg, false);
+		if (ret)
+			return ret;
+		return put_user(reg, uaddr);
+	}
+	}
+
+	return -ENXIO;
+}
+
+static int vgic_v2_has_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR:
+		switch (attr->attr) {
+		case KVM_VGIC_V2_ADDR_TYPE_DIST:
+		case KVM_VGIC_V2_ADDR_TYPE_CPU:
+			return 0;
+		}
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS:
+		return vgic_v2_has_attr_regs(dev, attr);
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS:
+		return 0;
+	case KVM_DEV_ARM_VGIC_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_CTRL_INIT:
+			return 0;
+		}
+	}
+	return -ENXIO;
+}
+
+struct kvm_device_ops kvm_arm_vgic_v2_ops = {
+	.name = "kvm-arm-vgic-v2",
+	.create = vgic_create,
+	.destroy = vgic_destroy,
+	.set_attr = vgic_v2_set_attr,
+	.get_attr = vgic_v2_get_attr,
+	.has_attr = vgic_v2_has_attr,
+};
+
+int vgic_v3_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr,
+		       struct vgic_reg_attr *reg_attr)
+{
+	unsigned long vgic_mpidr, mpidr_reg;
+
+	/*
+	 * For KVM_DEV_ARM_VGIC_GRP_DIST_REGS group,
+	 * attr might not hold MPIDR. Hence assume vcpu0.
+	 */
+	if (attr->group != KVM_DEV_ARM_VGIC_GRP_DIST_REGS) {
+		vgic_mpidr = (attr->attr & KVM_DEV_ARM_VGIC_V3_MPIDR_MASK) >>
+			      KVM_DEV_ARM_VGIC_V3_MPIDR_SHIFT;
+
+		mpidr_reg = VGIC_TO_MPIDR(vgic_mpidr);
+		reg_attr->vcpu = kvm_mpidr_to_vcpu(dev->kvm, mpidr_reg);
+	} else {
+		reg_attr->vcpu = kvm_get_vcpu(dev->kvm, 0);
+	}
+
+	if (!reg_attr->vcpu)
+		return -EINVAL;
+
+	reg_attr->addr = attr->attr & KVM_DEV_ARM_VGIC_OFFSET_MASK;
+
+	return 0;
+}
+
+/*
+ * vgic_v3_attr_regs_access - allows user space to access VGIC v3 state
+ *
+ * @dev:      kvm device handle
+ * @attr:     kvm device attribute
+ * @reg:      address the value is read or written
+ * @is_write: true if userspace is writing a register
+ */
+static int vgic_v3_attr_regs_access(struct kvm_device *dev,
+				    struct kvm_device_attr *attr,
+				    u64 *reg, bool is_write)
+{
+	struct vgic_reg_attr reg_attr;
+	gpa_t addr;
+	struct kvm_vcpu *vcpu;
+	int ret;
+	u32 tmp32;
+
+	ret = vgic_v3_parse_attr(dev, attr, &reg_attr);
+	if (ret)
+		return ret;
+
+	vcpu = reg_attr.vcpu;
+	addr = reg_attr.addr;
+
+	mutex_lock(&dev->kvm->lock);
+
+	if (unlikely(!vgic_initialized(dev->kvm))) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	if (!lock_all_vcpus(dev->kvm)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+		if (is_write)
+			tmp32 = *reg;
+
+		ret = vgic_v3_dist_uaccess(vcpu, is_write, addr, &tmp32);
+		if (!is_write)
+			*reg = tmp32;
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS:
+		if (is_write)
+			tmp32 = *reg;
+
+		ret = vgic_v3_redist_uaccess(vcpu, is_write, addr, &tmp32);
+		if (!is_write)
+			*reg = tmp32;
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: {
+		u64 regid;
+
+		regid = (attr->attr & KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK);
+		ret = vgic_v3_cpu_sysregs_uaccess(vcpu, is_write,
+						  regid, reg);
+		break;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: {
+		unsigned int info, intid;
+
+		info = (attr->attr & KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK) >>
+			KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT;
+		if (info == VGIC_LEVEL_INFO_LINE_LEVEL) {
+			intid = attr->attr &
+				KVM_DEV_ARM_VGIC_LINE_LEVEL_INTID_MASK;
+			ret = vgic_v3_line_level_info_uaccess(vcpu, is_write,
+							      intid, reg);
+		} else {
+			ret = -EINVAL;
+		}
+		break;
+	}
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	unlock_all_vcpus(dev->kvm);
+out:
+	mutex_unlock(&dev->kvm->lock);
+	return ret;
+}
+
+static int vgic_v3_set_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	int ret;
+
+	ret = vgic_set_common_attr(dev, attr);
+	if (ret != -ENXIO)
+		return ret;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u32 tmp32;
+		u64 reg;
+
+		if (get_user(tmp32, uaddr))
+			return -EFAULT;
+
+		reg = tmp32;
+		return vgic_v3_attr_regs_access(dev, attr, &reg, true);
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: {
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		u64 reg;
+
+		if (get_user(reg, uaddr))
+			return -EFAULT;
+
+		return vgic_v3_attr_regs_access(dev, attr, &reg, true);
+	}
+	case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u64 reg;
+		u32 tmp32;
+
+		if (get_user(tmp32, uaddr))
+			return -EFAULT;
+
+		reg = tmp32;
+		return vgic_v3_attr_regs_access(dev, attr, &reg, true);
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CTRL: {
+		int ret;
+
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES:
+			mutex_lock(&dev->kvm->lock);
+
+			if (!lock_all_vcpus(dev->kvm)) {
+				mutex_unlock(&dev->kvm->lock);
+				return -EBUSY;
+			}
+			ret = vgic_v3_save_pending_tables(dev->kvm);
+			unlock_all_vcpus(dev->kvm);
+			mutex_unlock(&dev->kvm->lock);
+			return ret;
+		}
+		break;
+	}
+	}
+	return -ENXIO;
+}
+
+static int vgic_v3_get_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	int ret;
+
+	ret = vgic_get_common_attr(dev, attr);
+	if (ret != -ENXIO)
+		return ret;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u64 reg;
+		u32 tmp32;
+
+		ret = vgic_v3_attr_regs_access(dev, attr, &reg, false);
+		if (ret)
+			return ret;
+		tmp32 = reg;
+		return put_user(tmp32, uaddr);
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: {
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		u64 reg;
+
+		ret = vgic_v3_attr_regs_access(dev, attr, &reg, false);
+		if (ret)
+			return ret;
+		return put_user(reg, uaddr);
+	}
+	case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: {
+		u32 __user *uaddr = (u32 __user *)(long)attr->addr;
+		u64 reg;
+		u32 tmp32;
+
+		ret = vgic_v3_attr_regs_access(dev, attr, &reg, false);
+		if (ret)
+			return ret;
+		tmp32 = reg;
+		return put_user(tmp32, uaddr);
+	}
+	}
+	return -ENXIO;
+}
+
+static int vgic_v3_has_attr(struct kvm_device *dev,
+			    struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR:
+		switch (attr->attr) {
+		case KVM_VGIC_V3_ADDR_TYPE_DIST:
+		case KVM_VGIC_V3_ADDR_TYPE_REDIST:
+		case KVM_VGIC_V3_ADDR_TYPE_REDIST_REGION:
+			return 0;
+		}
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS:
+	case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS:
+		return vgic_v3_has_attr_regs(dev, attr);
+	case KVM_DEV_ARM_VGIC_GRP_NR_IRQS:
+		return 0;
+	case KVM_DEV_ARM_VGIC_GRP_LEVEL_INFO: {
+		if (((attr->attr & KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_MASK) >>
+		      KVM_DEV_ARM_VGIC_LINE_LEVEL_INFO_SHIFT) ==
+		      VGIC_LEVEL_INFO_LINE_LEVEL)
+			return 0;
+		break;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_CTRL_INIT:
+			return 0;
+		case KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES:
+			return 0;
+		}
+	}
+	return -ENXIO;
+}
+
+struct kvm_device_ops kvm_arm_vgic_v3_ops = {
+	.name = "kvm-arm-vgic-v3",
+	.create = vgic_create,
+	.destroy = vgic_destroy,
+	.set_attr = vgic_v3_set_attr,
+	.get_attr = vgic_v3_get_attr,
+	.has_attr = vgic_v3_has_attr,
+};

virt/kvm/arm/vgic/vgic-mmio-v2.c

@@ -0,0 +1,546 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * VGICv2 MMIO handling functions
+ */
+
+#include <linux/irqchip/arm-gic.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <linux/nospec.h>
+
+#include <kvm/iodev.h>
+#include <kvm/arm_vgic.h>
+
+#include "vgic.h"
+#include "vgic-mmio.h"
+
+/*
+ * The Revision field in the IIDR have the following meanings:
+ *
+ * Revision 1: Report GICv2 interrupts as group 0 instead of group 1
+ * Revision 2: Interrupt groups are guest-configurable and signaled using
+ * 	       their configured groups.
+ */
+
+static unsigned long vgic_mmio_read_v2_misc(struct kvm_vcpu *vcpu,
+					    gpa_t addr, unsigned int len)
+{
+	struct vgic_dist *vgic = &vcpu->kvm->arch.vgic;
+	u32 value;
+
+	switch (addr & 0x0c) {
+	case GIC_DIST_CTRL:
+		value = vgic->enabled ? GICD_ENABLE : 0;
+		break;
+	case GIC_DIST_CTR:
+		value = vgic->nr_spis + VGIC_NR_PRIVATE_IRQS;
+		value = (value >> 5) - 1;
+		value |= (atomic_read(&vcpu->kvm->online_vcpus) - 1) << 5;
+		break;
+	case GIC_DIST_IIDR:
+		value = (PRODUCT_ID_KVM << GICD_IIDR_PRODUCT_ID_SHIFT) |
+			(vgic->implementation_rev << GICD_IIDR_REVISION_SHIFT) |
+			(IMPLEMENTER_ARM << GICD_IIDR_IMPLEMENTER_SHIFT);
+		break;
+	default:
+		return 0;
+	}
+
+	return value;
+}
+
+static void vgic_mmio_write_v2_misc(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len,
+				    unsigned long val)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	bool was_enabled = dist->enabled;
+
+	switch (addr & 0x0c) {
+	case GIC_DIST_CTRL:
+		dist->enabled = val & GICD_ENABLE;
+		if (!was_enabled && dist->enabled)
+			vgic_kick_vcpus(vcpu->kvm);
+		break;
+	case GIC_DIST_CTR:
+	case GIC_DIST_IIDR:
+		/* Nothing to do */
+		return;
+	}
+}
+
+static int vgic_mmio_uaccess_write_v2_misc(struct kvm_vcpu *vcpu,
+					   gpa_t addr, unsigned int len,
+					   unsigned long val)
+{
+	switch (addr & 0x0c) {
+	case GIC_DIST_IIDR:
+		if (val != vgic_mmio_read_v2_misc(vcpu, addr, len))
+			return -EINVAL;
+
+		/*
+		 * If we observe a write to GICD_IIDR we know that userspace
+		 * has been updated and has had a chance to cope with older
+		 * kernels (VGICv2 IIDR.Revision == 0) incorrectly reporting
+		 * interrupts as group 1, and therefore we now allow groups to
+		 * be user writable.  Doing this by default would break
+		 * migration from old kernels to new kernels with legacy
+		 * userspace.
+		 */
+		vcpu->kvm->arch.vgic.v2_groups_user_writable = true;
+		return 0;
+	}
+
+	vgic_mmio_write_v2_misc(vcpu, addr, len, val);
+	return 0;
+}
+
+static int vgic_mmio_uaccess_write_v2_group(struct kvm_vcpu *vcpu,
+					    gpa_t addr, unsigned int len,
+					    unsigned long val)
+{
+	if (vcpu->kvm->arch.vgic.v2_groups_user_writable)
+		vgic_mmio_write_group(vcpu, addr, len, val);
+
+	return 0;
+}
+
+static void vgic_mmio_write_sgir(struct kvm_vcpu *source_vcpu,
+				 gpa_t addr, unsigned int len,
+				 unsigned long val)
+{
+	int nr_vcpus = atomic_read(&source_vcpu->kvm->online_vcpus);
+	int intid = val & 0xf;
+	int targets = (val >> 16) & 0xff;
+	int mode = (val >> 24) & 0x03;
+	int c;
+	struct kvm_vcpu *vcpu;
+	unsigned long flags;
+
+	switch (mode) {
+	case 0x0:		/* as specified by targets */
+		break;
+	case 0x1:
+		targets = (1U << nr_vcpus) - 1;			/* all, ... */
+		targets &= ~(1U << source_vcpu->vcpu_id);	/* but self */
+		break;
+	case 0x2:		/* this very vCPU only */
+		targets = (1U << source_vcpu->vcpu_id);
+		break;
+	case 0x3:		/* reserved */
+		return;
+	}
+
+	kvm_for_each_vcpu(c, vcpu, source_vcpu->kvm) {
+		struct vgic_irq *irq;
+
+		if (!(targets & (1U << c)))
+			continue;
+
+		irq = vgic_get_irq(source_vcpu->kvm, vcpu, intid);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		irq->pending_latch = true;
+		irq->source |= 1U << source_vcpu->vcpu_id;
+
+		vgic_queue_irq_unlock(source_vcpu->kvm, irq, flags);
+		vgic_put_irq(source_vcpu->kvm, irq);
+	}
+}
+
+static unsigned long vgic_mmio_read_target(struct kvm_vcpu *vcpu,
+					   gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 8);
+	int i;
+	u64 val = 0;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		val |= (u64)irq->targets << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return val;
+}
+
+static void vgic_mmio_write_target(struct kvm_vcpu *vcpu,
+				   gpa_t addr, unsigned int len,
+				   unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 8);
+	u8 cpu_mask = GENMASK(atomic_read(&vcpu->kvm->online_vcpus) - 1, 0);
+	int i;
+	unsigned long flags;
+
+	/* GICD_ITARGETSR[0-7] are read-only */
+	if (intid < VGIC_NR_PRIVATE_IRQS)
+		return;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, NULL, intid + i);
+		int target;
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		irq->targets = (val >> (i * 8)) & cpu_mask;
+		target = irq->targets ? __ffs(irq->targets) : 0;
+		irq->target_vcpu = kvm_get_vcpu(vcpu->kvm, target);
+
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+static unsigned long vgic_mmio_read_sgipend(struct kvm_vcpu *vcpu,
+					    gpa_t addr, unsigned int len)
+{
+	u32 intid = addr & 0x0f;
+	int i;
+	u64 val = 0;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		val |= (u64)irq->source << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+	return val;
+}
+
+static void vgic_mmio_write_sgipendc(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	u32 intid = addr & 0x0f;
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		irq->source &= ~((val >> (i * 8)) & 0xff);
+		if (!irq->source)
+			irq->pending_latch = false;
+
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+static void vgic_mmio_write_sgipends(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	u32 intid = addr & 0x0f;
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		irq->source |= (val >> (i * 8)) & 0xff;
+
+		if (irq->source) {
+			irq->pending_latch = true;
+			vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+		} else {
+			raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		}
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+#define GICC_ARCH_VERSION_V2	0x2
+
+/* These are for userland accesses only, there is no guest-facing emulation. */
+static unsigned long vgic_mmio_read_vcpuif(struct kvm_vcpu *vcpu,
+					   gpa_t addr, unsigned int len)
+{
+	struct vgic_vmcr vmcr;
+	u32 val;
+
+	vgic_get_vmcr(vcpu, &vmcr);
+
+	switch (addr & 0xff) {
+	case GIC_CPU_CTRL:
+		val = vmcr.grpen0 << GIC_CPU_CTRL_EnableGrp0_SHIFT;
+		val |= vmcr.grpen1 << GIC_CPU_CTRL_EnableGrp1_SHIFT;
+		val |= vmcr.ackctl << GIC_CPU_CTRL_AckCtl_SHIFT;
+		val |= vmcr.fiqen << GIC_CPU_CTRL_FIQEn_SHIFT;
+		val |= vmcr.cbpr << GIC_CPU_CTRL_CBPR_SHIFT;
+		val |= vmcr.eoim << GIC_CPU_CTRL_EOImodeNS_SHIFT;
+
+		break;
+	case GIC_CPU_PRIMASK:
+		/*
+		 * Our KVM_DEV_TYPE_ARM_VGIC_V2 device ABI exports the
+		 * the PMR field as GICH_VMCR.VMPriMask rather than
+		 * GICC_PMR.Priority, so we expose the upper five bits of
+		 * priority mask to userspace using the lower bits in the
+		 * unsigned long.
+		 */
+		val = (vmcr.pmr & GICV_PMR_PRIORITY_MASK) >>
+			GICV_PMR_PRIORITY_SHIFT;
+		break;
+	case GIC_CPU_BINPOINT:
+		val = vmcr.bpr;
+		break;
+	case GIC_CPU_ALIAS_BINPOINT:
+		val = vmcr.abpr;
+		break;
+	case GIC_CPU_IDENT:
+		val = ((PRODUCT_ID_KVM << 20) |
+		       (GICC_ARCH_VERSION_V2 << 16) |
+		       IMPLEMENTER_ARM);
+		break;
+	default:
+		return 0;
+	}
+
+	return val;
+}
+
+static void vgic_mmio_write_vcpuif(struct kvm_vcpu *vcpu,
+				   gpa_t addr, unsigned int len,
+				   unsigned long val)
+{
+	struct vgic_vmcr vmcr;
+
+	vgic_get_vmcr(vcpu, &vmcr);
+
+	switch (addr & 0xff) {
+	case GIC_CPU_CTRL:
+		vmcr.grpen0 = !!(val & GIC_CPU_CTRL_EnableGrp0);
+		vmcr.grpen1 = !!(val & GIC_CPU_CTRL_EnableGrp1);
+		vmcr.ackctl = !!(val & GIC_CPU_CTRL_AckCtl);
+		vmcr.fiqen = !!(val & GIC_CPU_CTRL_FIQEn);
+		vmcr.cbpr = !!(val & GIC_CPU_CTRL_CBPR);
+		vmcr.eoim = !!(val & GIC_CPU_CTRL_EOImodeNS);
+
+		break;
+	case GIC_CPU_PRIMASK:
+		/*
+		 * Our KVM_DEV_TYPE_ARM_VGIC_V2 device ABI exports the
+		 * the PMR field as GICH_VMCR.VMPriMask rather than
+		 * GICC_PMR.Priority, so we expose the upper five bits of
+		 * priority mask to userspace using the lower bits in the
+		 * unsigned long.
+		 */
+		vmcr.pmr = (val << GICV_PMR_PRIORITY_SHIFT) &
+			GICV_PMR_PRIORITY_MASK;
+		break;
+	case GIC_CPU_BINPOINT:
+		vmcr.bpr = val;
+		break;
+	case GIC_CPU_ALIAS_BINPOINT:
+		vmcr.abpr = val;
+		break;
+	}
+
+	vgic_set_vmcr(vcpu, &vmcr);
+}
+
+static unsigned long vgic_mmio_read_apr(struct kvm_vcpu *vcpu,
+					gpa_t addr, unsigned int len)
+{
+	int n; /* which APRn is this */
+
+	n = (addr >> 2) & 0x3;
+
+	if (kvm_vgic_global_state.type == VGIC_V2) {
+		/* GICv2 hardware systems support max. 32 groups */
+		if (n != 0)
+			return 0;
+		return vcpu->arch.vgic_cpu.vgic_v2.vgic_apr;
+	} else {
+		struct vgic_v3_cpu_if *vgicv3 = &vcpu->arch.vgic_cpu.vgic_v3;
+
+		if (n > vgic_v3_max_apr_idx(vcpu))
+			return 0;
+
+		n = array_index_nospec(n, 4);
+
+		/* GICv3 only uses ICH_AP1Rn for memory mapped (GICv2) guests */
+		return vgicv3->vgic_ap1r[n];
+	}
+}
+
+static void vgic_mmio_write_apr(struct kvm_vcpu *vcpu,
+				gpa_t addr, unsigned int len,
+				unsigned long val)
+{
+	int n; /* which APRn is this */
+
+	n = (addr >> 2) & 0x3;
+
+	if (kvm_vgic_global_state.type == VGIC_V2) {
+		/* GICv2 hardware systems support max. 32 groups */
+		if (n != 0)
+			return;
+		vcpu->arch.vgic_cpu.vgic_v2.vgic_apr = val;
+	} else {
+		struct vgic_v3_cpu_if *vgicv3 = &vcpu->arch.vgic_cpu.vgic_v3;
+
+		if (n > vgic_v3_max_apr_idx(vcpu))
+			return;
+
+		n = array_index_nospec(n, 4);
+
+		/* GICv3 only uses ICH_AP1Rn for memory mapped (GICv2) guests */
+		vgicv3->vgic_ap1r[n] = val;
+	}
+}
+
+static const struct vgic_register_region vgic_v2_dist_registers[] = {
+	REGISTER_DESC_WITH_LENGTH_UACCESS(GIC_DIST_CTRL,
+		vgic_mmio_read_v2_misc, vgic_mmio_write_v2_misc,
+		NULL, vgic_mmio_uaccess_write_v2_misc,
+		12, VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_IGROUP,
+		vgic_mmio_read_group, vgic_mmio_write_group,
+		NULL, vgic_mmio_uaccess_write_v2_group, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_ENABLE_SET,
+		vgic_mmio_read_enable, vgic_mmio_write_senable, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_ENABLE_CLEAR,
+		vgic_mmio_read_enable, vgic_mmio_write_cenable, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_PENDING_SET,
+		vgic_mmio_read_pending, vgic_mmio_write_spending, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_PENDING_CLEAR,
+		vgic_mmio_read_pending, vgic_mmio_write_cpending, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_ACTIVE_SET,
+		vgic_mmio_read_active, vgic_mmio_write_sactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_sactive, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_ACTIVE_CLEAR,
+		vgic_mmio_read_active, vgic_mmio_write_cactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_cactive, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_PRI,
+		vgic_mmio_read_priority, vgic_mmio_write_priority, NULL, NULL,
+		8, VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_TARGET,
+		vgic_mmio_read_target, vgic_mmio_write_target, NULL, NULL, 8,
+		VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ(GIC_DIST_CONFIG,
+		vgic_mmio_read_config, vgic_mmio_write_config, NULL, NULL, 2,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_DIST_SOFTINT,
+		vgic_mmio_read_raz, vgic_mmio_write_sgir, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_DIST_SGI_PENDING_CLEAR,
+		vgic_mmio_read_sgipend, vgic_mmio_write_sgipendc, 16,
+		VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_DIST_SGI_PENDING_SET,
+		vgic_mmio_read_sgipend, vgic_mmio_write_sgipends, 16,
+		VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+};
+
+static const struct vgic_register_region vgic_v2_cpu_registers[] = {
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_CTRL,
+		vgic_mmio_read_vcpuif, vgic_mmio_write_vcpuif, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_PRIMASK,
+		vgic_mmio_read_vcpuif, vgic_mmio_write_vcpuif, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_BINPOINT,
+		vgic_mmio_read_vcpuif, vgic_mmio_write_vcpuif, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_ALIAS_BINPOINT,
+		vgic_mmio_read_vcpuif, vgic_mmio_write_vcpuif, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_ACTIVEPRIO,
+		vgic_mmio_read_apr, vgic_mmio_write_apr, 16,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GIC_CPU_IDENT,
+		vgic_mmio_read_vcpuif, vgic_mmio_write_vcpuif, 4,
+		VGIC_ACCESS_32bit),
+};
+
+unsigned int vgic_v2_init_dist_iodev(struct vgic_io_device *dev)
+{
+	dev->regions = vgic_v2_dist_registers;
+	dev->nr_regions = ARRAY_SIZE(vgic_v2_dist_registers);
+
+	kvm_iodevice_init(&dev->dev, &kvm_io_gic_ops);
+
+	return SZ_4K;
+}
+
+int vgic_v2_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr)
+{
+	const struct vgic_register_region *region;
+	struct vgic_io_device iodev;
+	struct vgic_reg_attr reg_attr;
+	struct kvm_vcpu *vcpu;
+	gpa_t addr;
+	int ret;
+
+	ret = vgic_v2_parse_attr(dev, attr, &reg_attr);
+	if (ret)
+		return ret;
+
+	vcpu = reg_attr.vcpu;
+	addr = reg_attr.addr;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+		iodev.regions = vgic_v2_dist_registers;
+		iodev.nr_regions = ARRAY_SIZE(vgic_v2_dist_registers);
+		iodev.base_addr = 0;
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_CPU_REGS:
+		iodev.regions = vgic_v2_cpu_registers;
+		iodev.nr_regions = ARRAY_SIZE(vgic_v2_cpu_registers);
+		iodev.base_addr = 0;
+		break;
+	default:
+		return -ENXIO;
+	}
+
+	/* We only support aligned 32-bit accesses. */
+	if (addr & 3)
+		return -ENXIO;
+
+	region = vgic_get_mmio_region(vcpu, &iodev, addr, sizeof(u32));
+	if (!region)
+		return -ENXIO;
+
+	return 0;
+}
+
+int vgic_v2_cpuif_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			  int offset, u32 *val)
+{
+	struct vgic_io_device dev = {
+		.regions = vgic_v2_cpu_registers,
+		.nr_regions = ARRAY_SIZE(vgic_v2_cpu_registers),
+		.iodev_type = IODEV_CPUIF,
+	};
+
+	return vgic_uaccess(vcpu, &dev, is_write, offset, val);
+}
+
+int vgic_v2_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 int offset, u32 *val)
+{
+	struct vgic_io_device dev = {
+		.regions = vgic_v2_dist_registers,
+		.nr_regions = ARRAY_SIZE(vgic_v2_dist_registers),
+		.iodev_type = IODEV_DIST,
+	};
+
+	return vgic_uaccess(vcpu, &dev, is_write, offset, val);
+}

virt/kvm/arm/vgic/vgic-mmio-v3.c

@@ -0,0 +1,986 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * VGICv3 MMIO handling functions
+ */
+
+#include <linux/irqchip/arm-gic-v3.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <kvm/iodev.h>
+#include <kvm/arm_vgic.h>
+
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_arm.h>
+#include <asm/kvm_mmu.h>
+
+#include "vgic.h"
+#include "vgic-mmio.h"
+
+/* extract @num bytes at @offset bytes offset in data */
+unsigned long extract_bytes(u64 data, unsigned int offset,
+			    unsigned int num)
+{
+	return (data >> (offset * 8)) & GENMASK_ULL(num * 8 - 1, 0);
+}
+
+/* allows updates of any half of a 64-bit register (or the whole thing) */
+u64 update_64bit_reg(u64 reg, unsigned int offset, unsigned int len,
+		     unsigned long val)
+{
+	int lower = (offset & 4) * 8;
+	int upper = lower + 8 * len - 1;
+
+	reg &= ~GENMASK_ULL(upper, lower);
+	val &= GENMASK_ULL(len * 8 - 1, 0);
+
+	return reg | ((u64)val << lower);
+}
+
+bool vgic_has_its(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+
+	if (dist->vgic_model != KVM_DEV_TYPE_ARM_VGIC_V3)
+		return false;
+
+	return dist->has_its;
+}
+
+bool vgic_supports_direct_msis(struct kvm *kvm)
+{
+	return kvm_vgic_global_state.has_gicv4 && vgic_has_its(kvm);
+}
+
+/*
+ * The Revision field in the IIDR have the following meanings:
+ *
+ * Revision 2: Interrupt groups are guest-configurable and signaled using
+ * 	       their configured groups.
+ */
+
+static unsigned long vgic_mmio_read_v3_misc(struct kvm_vcpu *vcpu,
+					    gpa_t addr, unsigned int len)
+{
+	struct vgic_dist *vgic = &vcpu->kvm->arch.vgic;
+	u32 value = 0;
+
+	switch (addr & 0x0c) {
+	case GICD_CTLR:
+		if (vgic->enabled)
+			value |= GICD_CTLR_ENABLE_SS_G1;
+		value |= GICD_CTLR_ARE_NS | GICD_CTLR_DS;
+		break;
+	case GICD_TYPER:
+		value = vgic->nr_spis + VGIC_NR_PRIVATE_IRQS;
+		value = (value >> 5) - 1;
+		if (vgic_has_its(vcpu->kvm)) {
+			value |= (INTERRUPT_ID_BITS_ITS - 1) << 19;
+			value |= GICD_TYPER_LPIS;
+		} else {
+			value |= (INTERRUPT_ID_BITS_SPIS - 1) << 19;
+		}
+		break;
+	case GICD_IIDR:
+		value = (PRODUCT_ID_KVM << GICD_IIDR_PRODUCT_ID_SHIFT) |
+			(vgic->implementation_rev << GICD_IIDR_REVISION_SHIFT) |
+			(IMPLEMENTER_ARM << GICD_IIDR_IMPLEMENTER_SHIFT);
+		break;
+	default:
+		return 0;
+	}
+
+	return value;
+}
+
+static void vgic_mmio_write_v3_misc(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len,
+				    unsigned long val)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	bool was_enabled = dist->enabled;
+
+	switch (addr & 0x0c) {
+	case GICD_CTLR:
+		dist->enabled = val & GICD_CTLR_ENABLE_SS_G1;
+
+		if (!was_enabled && dist->enabled)
+			vgic_kick_vcpus(vcpu->kvm);
+		break;
+	case GICD_TYPER:
+	case GICD_IIDR:
+		return;
+	}
+}
+
+static int vgic_mmio_uaccess_write_v3_misc(struct kvm_vcpu *vcpu,
+					   gpa_t addr, unsigned int len,
+					   unsigned long val)
+{
+	switch (addr & 0x0c) {
+	case GICD_IIDR:
+		if (val != vgic_mmio_read_v3_misc(vcpu, addr, len))
+			return -EINVAL;
+	}
+
+	vgic_mmio_write_v3_misc(vcpu, addr, len, val);
+	return 0;
+}
+
+static unsigned long vgic_mmio_read_irouter(struct kvm_vcpu *vcpu,
+					    gpa_t addr, unsigned int len)
+{
+	int intid = VGIC_ADDR_TO_INTID(addr, 64);
+	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, NULL, intid);
+	unsigned long ret = 0;
+
+	if (!irq)
+		return 0;
+
+	/* The upper word is RAZ for us. */
+	if (!(addr & 4))
+		ret = extract_bytes(READ_ONCE(irq->mpidr), addr & 7, len);
+
+	vgic_put_irq(vcpu->kvm, irq);
+	return ret;
+}
+
+static void vgic_mmio_write_irouter(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len,
+				    unsigned long val)
+{
+	int intid = VGIC_ADDR_TO_INTID(addr, 64);
+	struct vgic_irq *irq;
+	unsigned long flags;
+
+	/* The upper word is WI for us since we don't implement Aff3. */
+	if (addr & 4)
+		return;
+
+	irq = vgic_get_irq(vcpu->kvm, NULL, intid);
+
+	if (!irq)
+		return;
+
+	raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+	/* We only care about and preserve Aff0, Aff1 and Aff2. */
+	irq->mpidr = val & GENMASK(23, 0);
+	irq->target_vcpu = kvm_mpidr_to_vcpu(vcpu->kvm, irq->mpidr);
+
+	raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+	vgic_put_irq(vcpu->kvm, irq);
+}
+
+static unsigned long vgic_mmio_read_v3r_ctlr(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	return vgic_cpu->lpis_enabled ? GICR_CTLR_ENABLE_LPIS : 0;
+}
+
+
+static void vgic_mmio_write_v3r_ctlr(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	bool was_enabled = vgic_cpu->lpis_enabled;
+
+	if (!vgic_has_its(vcpu->kvm))
+		return;
+
+	vgic_cpu->lpis_enabled = val & GICR_CTLR_ENABLE_LPIS;
+
+	if (was_enabled && !vgic_cpu->lpis_enabled) {
+		vgic_flush_pending_lpis(vcpu);
+		vgic_its_invalidate_cache(vcpu->kvm);
+	}
+
+	if (!was_enabled && vgic_cpu->lpis_enabled)
+		vgic_enable_lpis(vcpu);
+}
+
+static unsigned long vgic_mmio_read_v3r_typer(struct kvm_vcpu *vcpu,
+					      gpa_t addr, unsigned int len)
+{
+	unsigned long mpidr = kvm_vcpu_get_mpidr_aff(vcpu);
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_redist_region *rdreg = vgic_cpu->rdreg;
+	int target_vcpu_id = vcpu->vcpu_id;
+	gpa_t last_rdist_typer = rdreg->base + GICR_TYPER +
+			(rdreg->free_index - 1) * KVM_VGIC_V3_REDIST_SIZE;
+	u64 value;
+
+	value = (u64)(mpidr & GENMASK(23, 0)) << 32;
+	value |= ((target_vcpu_id & 0xffff) << 8);
+
+	if (addr == last_rdist_typer)
+		value |= GICR_TYPER_LAST;
+	if (vgic_has_its(vcpu->kvm))
+		value |= GICR_TYPER_PLPIS;
+
+	return extract_bytes(value, addr & 7, len);
+}
+
+static unsigned long vgic_mmio_read_v3r_iidr(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	return (PRODUCT_ID_KVM << 24) | (IMPLEMENTER_ARM << 0);
+}
+
+static unsigned long vgic_mmio_read_v3_idregs(struct kvm_vcpu *vcpu,
+					      gpa_t addr, unsigned int len)
+{
+	switch (addr & 0xffff) {
+	case GICD_PIDR2:
+		/* report a GICv3 compliant implementation */
+		return 0x3b;
+	}
+
+	return 0;
+}
+
+static unsigned long vgic_v3_uaccess_read_pending(struct kvm_vcpu *vcpu,
+						  gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 value = 0;
+	int i;
+
+	/*
+	 * pending state of interrupt is latched in pending_latch variable.
+	 * Userspace will save and restore pending state and line_level
+	 * separately.
+	 * Refer to Documentation/virt/kvm/devices/arm-vgic-v3.txt
+	 * for handling of ISPENDR and ICPENDR.
+	 */
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		if (irq->pending_latch)
+			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+static int vgic_v3_uaccess_write_pending(struct kvm_vcpu *vcpu,
+					 gpa_t addr, unsigned int len,
+					 unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		if (test_bit(i, &val)) {
+			/*
+			 * pending_latch is set irrespective of irq type
+			 * (level or edge) to avoid dependency that VM should
+			 * restore irq config before pending info.
+			 */
+			irq->pending_latch = true;
+			vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+		} else {
+			irq->pending_latch = false;
+			raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		}
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return 0;
+}
+
+/* We want to avoid outer shareable. */
+u64 vgic_sanitise_shareability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_OuterShareable:
+		return GIC_BASER_InnerShareable;
+	default:
+		return field;
+	}
+}
+
+/* Avoid any inner non-cacheable mapping. */
+u64 vgic_sanitise_inner_cacheability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_CACHE_nCnB:
+	case GIC_BASER_CACHE_nC:
+		return GIC_BASER_CACHE_RaWb;
+	default:
+		return field;
+	}
+}
+
+/* Non-cacheable or same-as-inner are OK. */
+u64 vgic_sanitise_outer_cacheability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_CACHE_SameAsInner:
+	case GIC_BASER_CACHE_nC:
+		return field;
+	default:
+		return GIC_BASER_CACHE_nC;
+	}
+}
+
+u64 vgic_sanitise_field(u64 reg, u64 field_mask, int field_shift,
+			u64 (*sanitise_fn)(u64))
+{
+	u64 field = (reg & field_mask) >> field_shift;
+
+	field = sanitise_fn(field) << field_shift;
+	return (reg & ~field_mask) | field;
+}
+
+#define PROPBASER_RES0_MASK						\
+	(GENMASK_ULL(63, 59) | GENMASK_ULL(55, 52) | GENMASK_ULL(6, 5))
+#define PENDBASER_RES0_MASK						\
+	(BIT_ULL(63) | GENMASK_ULL(61, 59) | GENMASK_ULL(55, 52) |	\
+	 GENMASK_ULL(15, 12) | GENMASK_ULL(6, 0))
+
+static u64 vgic_sanitise_pendbaser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_SHAREABILITY_MASK,
+				  GICR_PENDBASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_INNER_CACHEABILITY_MASK,
+				  GICR_PENDBASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_OUTER_CACHEABILITY_MASK,
+				  GICR_PENDBASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	reg &= ~PENDBASER_RES0_MASK;
+
+	return reg;
+}
+
+static u64 vgic_sanitise_propbaser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_SHAREABILITY_MASK,
+				  GICR_PROPBASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_INNER_CACHEABILITY_MASK,
+				  GICR_PROPBASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_OUTER_CACHEABILITY_MASK,
+				  GICR_PROPBASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	reg &= ~PROPBASER_RES0_MASK;
+	return reg;
+}
+
+static unsigned long vgic_mmio_read_propbase(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	return extract_bytes(dist->propbaser, addr & 7, len);
+}
+
+static void vgic_mmio_write_propbase(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	u64 old_propbaser, propbaser;
+
+	/* Storing a value with LPIs already enabled is undefined */
+	if (vgic_cpu->lpis_enabled)
+		return;
+
+	do {
+		old_propbaser = READ_ONCE(dist->propbaser);
+		propbaser = old_propbaser;
+		propbaser = update_64bit_reg(propbaser, addr & 4, len, val);
+		propbaser = vgic_sanitise_propbaser(propbaser);
+	} while (cmpxchg64(&dist->propbaser, old_propbaser,
+			   propbaser) != old_propbaser);
+}
+
+static unsigned long vgic_mmio_read_pendbase(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	return extract_bytes(vgic_cpu->pendbaser, addr & 7, len);
+}
+
+static void vgic_mmio_write_pendbase(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	u64 old_pendbaser, pendbaser;
+
+	/* Storing a value with LPIs already enabled is undefined */
+	if (vgic_cpu->lpis_enabled)
+		return;
+
+	do {
+		old_pendbaser = READ_ONCE(vgic_cpu->pendbaser);
+		pendbaser = old_pendbaser;
+		pendbaser = update_64bit_reg(pendbaser, addr & 4, len, val);
+		pendbaser = vgic_sanitise_pendbaser(pendbaser);
+	} while (cmpxchg64(&vgic_cpu->pendbaser, old_pendbaser,
+			   pendbaser) != old_pendbaser);
+}
+
+/*
+ * The GICv3 per-IRQ registers are split to control PPIs and SGIs in the
+ * redistributors, while SPIs are covered by registers in the distributor
+ * block. Trying to set private IRQs in this block gets ignored.
+ * We take some special care here to fix the calculation of the register
+ * offset.
+ */
+#define REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(off, rd, wr, ur, uw, bpi, acc) \
+	{								\
+		.reg_offset = off,					\
+		.bits_per_irq = bpi,					\
+		.len = (bpi * VGIC_NR_PRIVATE_IRQS) / 8,		\
+		.access_flags = acc,					\
+		.read = vgic_mmio_read_raz,				\
+		.write = vgic_mmio_write_wi,				\
+	}, {								\
+		.reg_offset = off + (bpi * VGIC_NR_PRIVATE_IRQS) / 8,	\
+		.bits_per_irq = bpi,					\
+		.len = (bpi * (1024 - VGIC_NR_PRIVATE_IRQS)) / 8,	\
+		.access_flags = acc,					\
+		.read = rd,						\
+		.write = wr,						\
+		.uaccess_read = ur,					\
+		.uaccess_write = uw,					\
+	}
+
+static const struct vgic_register_region vgic_v3_dist_registers[] = {
+	REGISTER_DESC_WITH_LENGTH_UACCESS(GICD_CTLR,
+		vgic_mmio_read_v3_misc, vgic_mmio_write_v3_misc,
+		NULL, vgic_mmio_uaccess_write_v3_misc,
+		16, VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICD_STATUSR,
+		vgic_mmio_read_rao, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IGROUPR,
+		vgic_mmio_read_group, vgic_mmio_write_group, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISENABLER,
+		vgic_mmio_read_enable, vgic_mmio_write_senable, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICENABLER,
+		vgic_mmio_read_enable, vgic_mmio_write_cenable, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISPENDR,
+		vgic_mmio_read_pending, vgic_mmio_write_spending,
+		vgic_v3_uaccess_read_pending, vgic_v3_uaccess_write_pending, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICPENDR,
+		vgic_mmio_read_pending, vgic_mmio_write_cpending,
+		vgic_mmio_read_raz, vgic_mmio_uaccess_write_wi, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ISACTIVER,
+		vgic_mmio_read_active, vgic_mmio_write_sactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_sactive, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICACTIVER,
+		vgic_mmio_read_active, vgic_mmio_write_cactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_cactive,
+		1, VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IPRIORITYR,
+		vgic_mmio_read_priority, vgic_mmio_write_priority, NULL, NULL,
+		8, VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ITARGETSR,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, NULL, NULL, 8,
+		VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_ICFGR,
+		vgic_mmio_read_config, vgic_mmio_write_config, NULL, NULL, 2,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IGRPMODR,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, NULL, NULL, 1,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_BITS_PER_IRQ_SHARED(GICD_IROUTER,
+		vgic_mmio_read_irouter, vgic_mmio_write_irouter, NULL, NULL, 64,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICD_IDREGS,
+		vgic_mmio_read_v3_idregs, vgic_mmio_write_wi, 48,
+		VGIC_ACCESS_32bit),
+};
+
+static const struct vgic_register_region vgic_v3_rd_registers[] = {
+	/* RD_base registers */
+	REGISTER_DESC_WITH_LENGTH(GICR_CTLR,
+		vgic_mmio_read_v3r_ctlr, vgic_mmio_write_v3r_ctlr, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_STATUSR,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_IIDR,
+		vgic_mmio_read_v3r_iidr, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_TYPER,
+		vgic_mmio_read_v3r_typer, vgic_mmio_write_wi, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_WAKER,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_PROPBASER,
+		vgic_mmio_read_propbase, vgic_mmio_write_propbase, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_PENDBASER,
+		vgic_mmio_read_pendbase, vgic_mmio_write_pendbase, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(GICR_IDREGS,
+		vgic_mmio_read_v3_idregs, vgic_mmio_write_wi, 48,
+		VGIC_ACCESS_32bit),
+	/* SGI_base registers */
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_IGROUPR0,
+		vgic_mmio_read_group, vgic_mmio_write_group, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_ISENABLER0,
+		vgic_mmio_read_enable, vgic_mmio_write_senable, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_ICENABLER0,
+		vgic_mmio_read_enable, vgic_mmio_write_cenable, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH_UACCESS(SZ_64K + GICR_ISPENDR0,
+		vgic_mmio_read_pending, vgic_mmio_write_spending,
+		vgic_v3_uaccess_read_pending, vgic_v3_uaccess_write_pending, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH_UACCESS(SZ_64K + GICR_ICPENDR0,
+		vgic_mmio_read_pending, vgic_mmio_write_cpending,
+		vgic_mmio_read_raz, vgic_mmio_uaccess_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH_UACCESS(SZ_64K + GICR_ISACTIVER0,
+		vgic_mmio_read_active, vgic_mmio_write_sactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_sactive, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH_UACCESS(SZ_64K + GICR_ICACTIVER0,
+		vgic_mmio_read_active, vgic_mmio_write_cactive,
+		vgic_uaccess_read_active, vgic_mmio_uaccess_write_cactive, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_IPRIORITYR0,
+		vgic_mmio_read_priority, vgic_mmio_write_priority, 32,
+		VGIC_ACCESS_32bit | VGIC_ACCESS_8bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_ICFGR0,
+		vgic_mmio_read_config, vgic_mmio_write_config, 8,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_IGRPMODR0,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_DESC_WITH_LENGTH(SZ_64K + GICR_NSACR,
+		vgic_mmio_read_raz, vgic_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+};
+
+unsigned int vgic_v3_init_dist_iodev(struct vgic_io_device *dev)
+{
+	dev->regions = vgic_v3_dist_registers;
+	dev->nr_regions = ARRAY_SIZE(vgic_v3_dist_registers);
+
+	kvm_iodevice_init(&dev->dev, &kvm_io_gic_ops);
+
+	return SZ_64K;
+}
+
+/**
+ * vgic_register_redist_iodev - register a single redist iodev
+ * @vcpu:    The VCPU to which the redistributor belongs
+ *
+ * Register a KVM iodev for this VCPU's redistributor using the address
+ * provided.
+ *
+ * Return 0 on success, -ERRNO otherwise.
+ */
+int vgic_register_redist_iodev(struct kvm_vcpu *vcpu)
+{
+	struct kvm *kvm = vcpu->kvm;
+	struct vgic_dist *vgic = &kvm->arch.vgic;
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_io_device *rd_dev = &vcpu->arch.vgic_cpu.rd_iodev;
+	struct vgic_redist_region *rdreg;
+	gpa_t rd_base;
+	int ret;
+
+	if (!IS_VGIC_ADDR_UNDEF(vgic_cpu->rd_iodev.base_addr))
+		return 0;
+
+	/*
+	 * We may be creating VCPUs before having set the base address for the
+	 * redistributor region, in which case we will come back to this
+	 * function for all VCPUs when the base address is set.  Just return
+	 * without doing any work for now.
+	 */
+	rdreg = vgic_v3_rdist_free_slot(&vgic->rd_regions);
+	if (!rdreg)
+		return 0;
+
+	if (!vgic_v3_check_base(kvm))
+		return -EINVAL;
+
+	vgic_cpu->rdreg = rdreg;
+
+	rd_base = rdreg->base + rdreg->free_index * KVM_VGIC_V3_REDIST_SIZE;
+
+	kvm_iodevice_init(&rd_dev->dev, &kvm_io_gic_ops);
+	rd_dev->base_addr = rd_base;
+	rd_dev->iodev_type = IODEV_REDIST;
+	rd_dev->regions = vgic_v3_rd_registers;
+	rd_dev->nr_regions = ARRAY_SIZE(vgic_v3_rd_registers);
+	rd_dev->redist_vcpu = vcpu;
+
+	mutex_lock(&kvm->slots_lock);
+	ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, rd_base,
+				      2 * SZ_64K, &rd_dev->dev);
+	mutex_unlock(&kvm->slots_lock);
+
+	if (ret)
+		return ret;
+
+	rdreg->free_index++;
+	return 0;
+}
+
+static void vgic_unregister_redist_iodev(struct kvm_vcpu *vcpu)
+{
+	struct vgic_io_device *rd_dev = &vcpu->arch.vgic_cpu.rd_iodev;
+
+	kvm_io_bus_unregister_dev(vcpu->kvm, KVM_MMIO_BUS, &rd_dev->dev);
+}
+
+static int vgic_register_all_redist_iodevs(struct kvm *kvm)
+{
+	struct kvm_vcpu *vcpu;
+	int c, ret = 0;
+
+	kvm_for_each_vcpu(c, vcpu, kvm) {
+		ret = vgic_register_redist_iodev(vcpu);
+		if (ret)
+			break;
+	}
+
+	if (ret) {
+		/* The current c failed, so we start with the previous one. */
+		mutex_lock(&kvm->slots_lock);
+		for (c--; c >= 0; c--) {
+			vcpu = kvm_get_vcpu(kvm, c);
+			vgic_unregister_redist_iodev(vcpu);
+		}
+		mutex_unlock(&kvm->slots_lock);
+	}
+
+	return ret;
+}
+
+/**
+ * vgic_v3_insert_redist_region - Insert a new redistributor region
+ *
+ * Performs various checks before inserting the rdist region in the list.
+ * Those tests depend on whether the size of the rdist region is known
+ * (ie. count != 0). The list is sorted by rdist region index.
+ *
+ * @kvm: kvm handle
+ * @index: redist region index
+ * @base: base of the new rdist region
+ * @count: number of redistributors the region is made of (0 in the old style
+ * single region, whose size is induced from the number of vcpus)
+ *
+ * Return 0 on success, < 0 otherwise
+ */
+static int vgic_v3_insert_redist_region(struct kvm *kvm, uint32_t index,
+					gpa_t base, uint32_t count)
+{
+	struct vgic_dist *d = &kvm->arch.vgic;
+	struct vgic_redist_region *rdreg;
+	struct list_head *rd_regions = &d->rd_regions;
+	size_t size = count * KVM_VGIC_V3_REDIST_SIZE;
+	int ret;
+
+	/* single rdist region already set ?*/
+	if (!count && !list_empty(rd_regions))
+		return -EINVAL;
+
+	/* cross the end of memory ? */
+	if (base + size < base)
+		return -EINVAL;
+
+	if (list_empty(rd_regions)) {
+		if (index != 0)
+			return -EINVAL;
+	} else {
+		rdreg = list_last_entry(rd_regions,
+					struct vgic_redist_region, list);
+		if (index != rdreg->index + 1)
+			return -EINVAL;
+
+		/* Cannot add an explicitly sized regions after legacy region */
+		if (!rdreg->count)
+			return -EINVAL;
+	}
+
+	/*
+	 * For legacy single-region redistributor regions (!count),
+	 * check that the redistributor region does not overlap with the
+	 * distributor's address space.
+	 */
+	if (!count && !IS_VGIC_ADDR_UNDEF(d->vgic_dist_base) &&
+		vgic_dist_overlap(kvm, base, size))
+		return -EINVAL;
+
+	/* collision with any other rdist region? */
+	if (vgic_v3_rdist_overlap(kvm, base, size))
+		return -EINVAL;
+
+	rdreg = kzalloc(sizeof(*rdreg), GFP_KERNEL);
+	if (!rdreg)
+		return -ENOMEM;
+
+	rdreg->base = VGIC_ADDR_UNDEF;
+
+	ret = vgic_check_ioaddr(kvm, &rdreg->base, base, SZ_64K);
+	if (ret)
+		goto free;
+
+	rdreg->base = base;
+	rdreg->count = count;
+	rdreg->free_index = 0;
+	rdreg->index = index;
+
+	list_add_tail(&rdreg->list, rd_regions);
+	return 0;
+free:
+	kfree(rdreg);
+	return ret;
+}
+
+int vgic_v3_set_redist_base(struct kvm *kvm, u32 index, u64 addr, u32 count)
+{
+	int ret;
+
+	ret = vgic_v3_insert_redist_region(kvm, index, addr, count);
+	if (ret)
+		return ret;
+
+	/*
+	 * Register iodevs for each existing VCPU.  Adding more VCPUs
+	 * afterwards will register the iodevs when needed.
+	 */
+	ret = vgic_register_all_redist_iodevs(kvm);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+int vgic_v3_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr)
+{
+	const struct vgic_register_region *region;
+	struct vgic_io_device iodev;
+	struct vgic_reg_attr reg_attr;
+	struct kvm_vcpu *vcpu;
+	gpa_t addr;
+	int ret;
+
+	ret = vgic_v3_parse_attr(dev, attr, &reg_attr);
+	if (ret)
+		return ret;
+
+	vcpu = reg_attr.vcpu;
+	addr = reg_attr.addr;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_DIST_REGS:
+		iodev.regions = vgic_v3_dist_registers;
+		iodev.nr_regions = ARRAY_SIZE(vgic_v3_dist_registers);
+		iodev.base_addr = 0;
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_REDIST_REGS:{
+		iodev.regions = vgic_v3_rd_registers;
+		iodev.nr_regions = ARRAY_SIZE(vgic_v3_rd_registers);
+		iodev.base_addr = 0;
+		break;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS: {
+		u64 reg, id;
+
+		id = (attr->attr & KVM_DEV_ARM_VGIC_SYSREG_INSTR_MASK);
+		return vgic_v3_has_cpu_sysregs_attr(vcpu, 0, id, &reg);
+	}
+	default:
+		return -ENXIO;
+	}
+
+	/* We only support aligned 32-bit accesses. */
+	if (addr & 3)
+		return -ENXIO;
+
+	region = vgic_get_mmio_region(vcpu, &iodev, addr, sizeof(u32));
+	if (!region)
+		return -ENXIO;
+
+	return 0;
+}
+/*
+ * Compare a given affinity (level 1-3 and a level 0 mask, from the SGI
+ * generation register ICC_SGI1R_EL1) with a given VCPU.
+ * If the VCPU's MPIDR matches, return the level0 affinity, otherwise
+ * return -1.
+ */
+static int match_mpidr(u64 sgi_aff, u16 sgi_cpu_mask, struct kvm_vcpu *vcpu)
+{
+	unsigned long affinity;
+	int level0;
+
+	/*
+	 * Split the current VCPU's MPIDR into affinity level 0 and the
+	 * rest as this is what we have to compare against.
+	 */
+	affinity = kvm_vcpu_get_mpidr_aff(vcpu);
+	level0 = MPIDR_AFFINITY_LEVEL(affinity, 0);
+	affinity &= ~MPIDR_LEVEL_MASK;
+
+	/* bail out if the upper three levels don't match */
+	if (sgi_aff != affinity)
+		return -1;
+
+	/* Is this VCPU's bit set in the mask ? */
+	if (!(sgi_cpu_mask & BIT(level0)))
+		return -1;
+
+	return level0;
+}
+
+/*
+ * The ICC_SGI* registers encode the affinity differently from the MPIDR,
+ * so provide a wrapper to use the existing defines to isolate a certain
+ * affinity level.
+ */
+#define SGI_AFFINITY_LEVEL(reg, level) \
+	((((reg) & ICC_SGI1R_AFFINITY_## level ##_MASK) \
+	>> ICC_SGI1R_AFFINITY_## level ##_SHIFT) << MPIDR_LEVEL_SHIFT(level))
+
+/**
+ * vgic_v3_dispatch_sgi - handle SGI requests from VCPUs
+ * @vcpu: The VCPU requesting a SGI
+ * @reg: The value written into ICC_{ASGI1,SGI0,SGI1}R by that VCPU
+ * @allow_group1: Does the sysreg access allow generation of G1 SGIs
+ *
+ * With GICv3 (and ARE=1) CPUs trigger SGIs by writing to a system register.
+ * This will trap in sys_regs.c and call this function.
+ * This ICC_SGI1R_EL1 register contains the upper three affinity levels of the
+ * target processors as well as a bitmask of 16 Aff0 CPUs.
+ * If the interrupt routing mode bit is not set, we iterate over all VCPUs to
+ * check for matching ones. If this bit is set, we signal all, but not the
+ * calling VCPU.
+ */
+void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg, bool allow_group1)
+{
+	struct kvm *kvm = vcpu->kvm;
+	struct kvm_vcpu *c_vcpu;
+	u16 target_cpus;
+	u64 mpidr;
+	int sgi, c;
+	int vcpu_id = vcpu->vcpu_id;
+	bool broadcast;
+	unsigned long flags;
+
+	sgi = (reg & ICC_SGI1R_SGI_ID_MASK) >> ICC_SGI1R_SGI_ID_SHIFT;
+	broadcast = reg & BIT_ULL(ICC_SGI1R_IRQ_ROUTING_MODE_BIT);
+	target_cpus = (reg & ICC_SGI1R_TARGET_LIST_MASK) >> ICC_SGI1R_TARGET_LIST_SHIFT;
+	mpidr = SGI_AFFINITY_LEVEL(reg, 3);
+	mpidr |= SGI_AFFINITY_LEVEL(reg, 2);
+	mpidr |= SGI_AFFINITY_LEVEL(reg, 1);
+
+	/*
+	 * We iterate over all VCPUs to find the MPIDRs matching the request.
+	 * If we have handled one CPU, we clear its bit to detect early
+	 * if we are already finished. This avoids iterating through all
+	 * VCPUs when most of the times we just signal a single VCPU.
+	 */
+	kvm_for_each_vcpu(c, c_vcpu, kvm) {
+		struct vgic_irq *irq;
+
+		/* Exit early if we have dealt with all requested CPUs */
+		if (!broadcast && target_cpus == 0)
+			break;
+
+		/* Don't signal the calling VCPU */
+		if (broadcast && c == vcpu_id)
+			continue;
+
+		if (!broadcast) {
+			int level0;
+
+			level0 = match_mpidr(mpidr, target_cpus, c_vcpu);
+			if (level0 == -1)
+				continue;
+
+			/* remove this matching VCPU from the mask */
+			target_cpus &= ~BIT(level0);
+		}
+
+		irq = vgic_get_irq(vcpu->kvm, c_vcpu, sgi);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		/*
+		 * An access targetting Group0 SGIs can only generate
+		 * those, while an access targetting Group1 SGIs can
+		 * generate interrupts of either group.
+		 */
+		if (!irq->group || allow_group1) {
+			irq->pending_latch = true;
+			vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+		} else {
+			raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		}
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+int vgic_v3_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 int offset, u32 *val)
+{
+	struct vgic_io_device dev = {
+		.regions = vgic_v3_dist_registers,
+		.nr_regions = ARRAY_SIZE(vgic_v3_dist_registers),
+	};
+
+	return vgic_uaccess(vcpu, &dev, is_write, offset, val);
+}
+
+int vgic_v3_redist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			   int offset, u32 *val)
+{
+	struct vgic_io_device rd_dev = {
+		.regions = vgic_v3_rd_registers,
+		.nr_regions = ARRAY_SIZE(vgic_v3_rd_registers),
+	};
+
+	return vgic_uaccess(vcpu, &rd_dev, is_write, offset, val);
+}
+
+int vgic_v3_line_level_info_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+				    u32 intid, u64 *val)
+{
+	if (intid % 32)
+		return -EINVAL;
+
+	if (is_write)
+		vgic_write_irq_line_level_info(vcpu, intid, *val);
+	else
+		*val = vgic_read_irq_line_level_info(vcpu, intid);
+
+	return 0;
+}

virt/kvm/arm/vgic/vgic-mmio.c

@@ -0,0 +1,947 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * VGIC MMIO handling functions
+ */
+
+#include <linux/bitops.h>
+#include <linux/bsearch.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <kvm/iodev.h>
+#include <kvm/arm_arch_timer.h>
+#include <kvm/arm_vgic.h>
+
+#include "vgic.h"
+#include "vgic-mmio.h"
+
+unsigned long vgic_mmio_read_raz(struct kvm_vcpu *vcpu,
+				 gpa_t addr, unsigned int len)
+{
+	return 0;
+}
+
+unsigned long vgic_mmio_read_rao(struct kvm_vcpu *vcpu,
+				 gpa_t addr, unsigned int len)
+{
+	return -1UL;
+}
+
+void vgic_mmio_write_wi(struct kvm_vcpu *vcpu, gpa_t addr,
+			unsigned int len, unsigned long val)
+{
+	/* Ignore */
+}
+
+int vgic_mmio_uaccess_write_wi(struct kvm_vcpu *vcpu, gpa_t addr,
+			       unsigned int len, unsigned long val)
+{
+	/* Ignore */
+	return 0;
+}
+
+unsigned long vgic_mmio_read_group(struct kvm_vcpu *vcpu,
+				   gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 value = 0;
+	int i;
+
+	/* Loop over all IRQs affected by this read */
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		if (irq->group)
+			value |= BIT(i);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+void vgic_mmio_write_group(struct kvm_vcpu *vcpu, gpa_t addr,
+			   unsigned int len, unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		irq->group = !!(val & BIT(i));
+		vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+/*
+ * Read accesses to both GICD_ICENABLER and GICD_ISENABLER return the value
+ * of the enabled bit, so there is only one function for both here.
+ */
+unsigned long vgic_mmio_read_enable(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 value = 0;
+	int i;
+
+	/* Loop over all IRQs affected by this read */
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		if (irq->enabled)
+			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+void vgic_mmio_write_senable(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		if (vgic_irq_is_mapped_level(irq)) {
+			bool was_high = irq->line_level;
+
+			/*
+			 * We need to update the state of the interrupt because
+			 * the guest might have changed the state of the device
+			 * while the interrupt was disabled at the VGIC level.
+			 */
+			irq->line_level = vgic_get_phys_line_level(irq);
+			/*
+			 * Deactivate the physical interrupt so the GIC will let
+			 * us know when it is asserted again.
+			 */
+			if (!irq->active && was_high && !irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+		irq->enabled = true;
+		vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+void vgic_mmio_write_cenable(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		irq->enabled = false;
+
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+unsigned long vgic_mmio_read_pending(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 value = 0;
+	int i;
+
+	/* Loop over all IRQs affected by this read */
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		unsigned long flags;
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		if (irq_is_pending(irq))
+			value |= (1U << i);
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+/*
+ * This function will return the VCPU that performed the MMIO access and
+ * trapped from within the VM, and will return NULL if this is a userspace
+ * access.
+ *
+ * We can disable preemption locally around accessing the per-CPU variable,
+ * and use the resolved vcpu pointer after enabling preemption again, because
+ * even if the current thread is migrated to another CPU, reading the per-CPU
+ * value later will give us the same value as we update the per-CPU variable
+ * in the preempt notifier handlers.
+ */
+static struct kvm_vcpu *vgic_get_mmio_requester_vcpu(void)
+{
+	struct kvm_vcpu *vcpu;
+
+	preempt_disable();
+	vcpu = kvm_arm_get_running_vcpu();
+	preempt_enable();
+	return vcpu;
+}
+
+/* Must be called with irq->irq_lock held */
+static void vgic_hw_irq_spending(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+				 bool is_uaccess)
+{
+	if (is_uaccess)
+		return;
+
+	irq->pending_latch = true;
+	vgic_irq_set_phys_active(irq, true);
+}
+
+static bool is_vgic_v2_sgi(struct kvm_vcpu *vcpu, struct vgic_irq *irq)
+{
+	return (vgic_irq_is_sgi(irq->intid) &&
+		vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V2);
+}
+
+void vgic_mmio_write_spending(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val)
+{
+	bool is_uaccess = !vgic_get_mmio_requester_vcpu();
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		/* GICD_ISPENDR0 SGI bits are WI */
+		if (is_vgic_v2_sgi(vcpu, irq)) {
+			vgic_put_irq(vcpu->kvm, irq);
+			continue;
+		}
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		if (irq->hw)
+			vgic_hw_irq_spending(vcpu, irq, is_uaccess);
+		else
+			irq->pending_latch = true;
+		vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+/* Must be called with irq->irq_lock held */
+static void vgic_hw_irq_cpending(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+				 bool is_uaccess)
+{
+	if (is_uaccess)
+		return;
+
+	irq->pending_latch = false;
+
+	/*
+	 * We don't want the guest to effectively mask the physical
+	 * interrupt by doing a write to SPENDR followed by a write to
+	 * CPENDR for HW interrupts, so we clear the active state on
+	 * the physical side if the virtual interrupt is not active.
+	 * This may lead to taking an additional interrupt on the
+	 * host, but that should not be a problem as the worst that
+	 * can happen is an additional vgic injection.  We also clear
+	 * the pending state to maintain proper semantics for edge HW
+	 * interrupts.
+	 */
+	vgic_irq_set_phys_pending(irq, false);
+	if (!irq->active)
+		vgic_irq_set_phys_active(irq, false);
+}
+
+void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val)
+{
+	bool is_uaccess = !vgic_get_mmio_requester_vcpu();
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+	unsigned long flags;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		/* GICD_ICPENDR0 SGI bits are WI */
+		if (is_vgic_v2_sgi(vcpu, irq)) {
+			vgic_put_irq(vcpu->kvm, irq);
+			continue;
+		}
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		if (irq->hw)
+			vgic_hw_irq_cpending(vcpu, irq, is_uaccess);
+		else
+			irq->pending_latch = false;
+
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+
+/*
+ * If we are fiddling with an IRQ's active state, we have to make sure the IRQ
+ * is not queued on some running VCPU's LRs, because then the change to the
+ * active state can be overwritten when the VCPU's state is synced coming back
+ * from the guest.
+ *
+ * For shared interrupts as well as GICv3 private interrupts, we have to
+ * stop all the VCPUs because interrupts can be migrated while we don't hold
+ * the IRQ locks and we don't want to be chasing moving targets.
+ *
+ * For GICv2 private interrupts we don't have to do anything because
+ * userspace accesses to the VGIC state already require all VCPUs to be
+ * stopped, and only the VCPU itself can modify its private interrupts
+ * active state, which guarantees that the VCPU is not running.
+ */
+static void vgic_access_active_prepare(struct kvm_vcpu *vcpu, u32 intid)
+{
+	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3 ||
+	    intid >= VGIC_NR_PRIVATE_IRQS)
+		kvm_arm_halt_guest(vcpu->kvm);
+}
+
+/* See vgic_access_active_prepare */
+static void vgic_access_active_finish(struct kvm_vcpu *vcpu, u32 intid)
+{
+	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3 ||
+	    intid >= VGIC_NR_PRIVATE_IRQS)
+		kvm_arm_resume_guest(vcpu->kvm);
+}
+
+static unsigned long __vgic_mmio_read_active(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 value = 0;
+	int i;
+
+	/* Loop over all IRQs affected by this read */
+	for (i = 0; i < len * 8; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		/*
+		 * Even for HW interrupts, don't evaluate the HW state as
+		 * all the guest is interested in is the virtual state.
+		 */
+		if (irq->active)
+			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+unsigned long vgic_mmio_read_active(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	u32 val;
+
+	mutex_lock(&vcpu->kvm->lock);
+	vgic_access_active_prepare(vcpu, intid);
+
+	val = __vgic_mmio_read_active(vcpu, addr, len);
+
+	vgic_access_active_finish(vcpu, intid);
+	mutex_unlock(&vcpu->kvm->lock);
+
+	return val;
+}
+
+unsigned long vgic_uaccess_read_active(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len)
+{
+	return __vgic_mmio_read_active(vcpu, addr, len);
+}
+
+/* Must be called with irq->irq_lock held */
+static void vgic_hw_irq_change_active(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+				      bool active, bool is_uaccess)
+{
+	if (is_uaccess)
+		return;
+
+	irq->active = active;
+	vgic_irq_set_phys_active(irq, active);
+}
+
+static void vgic_mmio_change_active(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+				    bool active)
+{
+	unsigned long flags;
+	struct kvm_vcpu *requester_vcpu = vgic_get_mmio_requester_vcpu();
+
+	raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+	if (irq->hw) {
+		vgic_hw_irq_change_active(vcpu, irq, active, !requester_vcpu);
+	} else {
+		u32 model = vcpu->kvm->arch.vgic.vgic_model;
+		u8 active_source;
+
+		irq->active = active;
+
+		/*
+		 * The GICv2 architecture indicates that the source CPUID for
+		 * an SGI should be provided during an EOI which implies that
+		 * the active state is stored somewhere, but at the same time
+		 * this state is not architecturally exposed anywhere and we
+		 * have no way of knowing the right source.
+		 *
+		 * This may lead to a VCPU not being able to receive
+		 * additional instances of a particular SGI after migration
+		 * for a GICv2 VM on some GIC implementations.  Oh well.
+		 */
+		active_source = (requester_vcpu) ? requester_vcpu->vcpu_id : 0;
+
+		if (model == KVM_DEV_TYPE_ARM_VGIC_V2 &&
+		    active && vgic_irq_is_sgi(irq->intid))
+			irq->active_source = active_source;
+	}
+
+	if (irq->active)
+		vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+	else
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+}
+
+static void __vgic_mmio_write_cactive(struct kvm_vcpu *vcpu,
+				      gpa_t addr, unsigned int len,
+				      unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		vgic_mmio_change_active(vcpu, irq, false);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+void vgic_mmio_write_cactive(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+
+	mutex_lock(&vcpu->kvm->lock);
+	vgic_access_active_prepare(vcpu, intid);
+
+	__vgic_mmio_write_cactive(vcpu, addr, len, val);
+
+	vgic_access_active_finish(vcpu, intid);
+	mutex_unlock(&vcpu->kvm->lock);
+}
+
+int vgic_mmio_uaccess_write_cactive(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	__vgic_mmio_write_cactive(vcpu, addr, len, val);
+	return 0;
+}
+
+static void __vgic_mmio_write_sactive(struct kvm_vcpu *vcpu,
+				      gpa_t addr, unsigned int len,
+				      unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+	int i;
+
+	for_each_set_bit(i, &val, len * 8) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		vgic_mmio_change_active(vcpu, irq, true);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+void vgic_mmio_write_sactive(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 1);
+
+	mutex_lock(&vcpu->kvm->lock);
+	vgic_access_active_prepare(vcpu, intid);
+
+	__vgic_mmio_write_sactive(vcpu, addr, len, val);
+
+	vgic_access_active_finish(vcpu, intid);
+	mutex_unlock(&vcpu->kvm->lock);
+}
+
+int vgic_mmio_uaccess_write_sactive(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	__vgic_mmio_write_sactive(vcpu, addr, len, val);
+	return 0;
+}
+
+unsigned long vgic_mmio_read_priority(struct kvm_vcpu *vcpu,
+				      gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 8);
+	int i;
+	u64 val = 0;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		val |= (u64)irq->priority << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return val;
+}
+
+/*
+ * We currently don't handle changing the priority of an interrupt that
+ * is already pending on a VCPU. If there is a need for this, we would
+ * need to make this VCPU exit and re-evaluate the priorities, potentially
+ * leading to this interrupt getting presented now to the guest (if it has
+ * been masked by the priority mask before).
+ */
+void vgic_mmio_write_priority(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 8);
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		/* Narrow the priority range to what we actually support */
+		irq->priority = (val >> (i * 8)) & GENMASK(7, 8 - VGIC_PRI_BITS);
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+unsigned long vgic_mmio_read_config(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 2);
+	u32 value = 0;
+	int i;
+
+	for (i = 0; i < len * 4; i++) {
+		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		if (irq->config == VGIC_CONFIG_EDGE)
+			value |= (2U << (i * 2));
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return value;
+}
+
+void vgic_mmio_write_config(struct kvm_vcpu *vcpu,
+			    gpa_t addr, unsigned int len,
+			    unsigned long val)
+{
+	u32 intid = VGIC_ADDR_TO_INTID(addr, 2);
+	int i;
+	unsigned long flags;
+
+	for (i = 0; i < len * 4; i++) {
+		struct vgic_irq *irq;
+
+		/*
+		 * The configuration cannot be changed for SGIs in general,
+		 * for PPIs this is IMPLEMENTATION DEFINED. The arch timer
+		 * code relies on PPIs being level triggered, so we also
+		 * make them read-only here.
+		 */
+		if (intid + i < VGIC_NR_PRIVATE_IRQS)
+			continue;
+
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+
+		if (test_bit(i * 2 + 1, &val))
+			irq->config = VGIC_CONFIG_EDGE;
+		else
+			irq->config = VGIC_CONFIG_LEVEL;
+
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+u64 vgic_read_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid)
+{
+	int i;
+	u64 val = 0;
+	int nr_irqs = vcpu->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS;
+
+	for (i = 0; i < 32; i++) {
+		struct vgic_irq *irq;
+
+		if ((intid + i) < VGIC_NR_SGIS || (intid + i) >= nr_irqs)
+			continue;
+
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		if (irq->config == VGIC_CONFIG_LEVEL && irq->line_level)
+			val |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	return val;
+}
+
+void vgic_write_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid,
+				    const u64 val)
+{
+	int i;
+	int nr_irqs = vcpu->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS;
+	unsigned long flags;
+
+	for (i = 0; i < 32; i++) {
+		struct vgic_irq *irq;
+		bool new_level;
+
+		if ((intid + i) < VGIC_NR_SGIS || (intid + i) >= nr_irqs)
+			continue;
+
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+
+		/*
+		 * Line level is set irrespective of irq type
+		 * (level or edge) to avoid dependency that VM should
+		 * restore irq config before line level.
+		 */
+		new_level = !!(val & (1U << i));
+		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		irq->line_level = new_level;
+		if (new_level)
+			vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+		else
+			raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+}
+
+static int match_region(const void *key, const void *elt)
+{
+	const unsigned int offset = (unsigned long)key;
+	const struct vgic_register_region *region = elt;
+
+	if (offset < region->reg_offset)
+		return -1;
+
+	if (offset >= region->reg_offset + region->len)
+		return 1;
+
+	return 0;
+}
+
+const struct vgic_register_region *
+vgic_find_mmio_region(const struct vgic_register_region *regions,
+		      int nr_regions, unsigned int offset)
+{
+	return bsearch((void *)(uintptr_t)offset, regions, nr_regions,
+		       sizeof(regions[0]), match_region);
+}
+
+void vgic_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr)
+{
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_set_vmcr(vcpu, vmcr);
+	else
+		vgic_v3_set_vmcr(vcpu, vmcr);
+}
+
+void vgic_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr)
+{
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_get_vmcr(vcpu, vmcr);
+	else
+		vgic_v3_get_vmcr(vcpu, vmcr);
+}
+
+/*
+ * kvm_mmio_read_buf() returns a value in a format where it can be converted
+ * to a byte array and be directly observed as the guest wanted it to appear
+ * in memory if it had done the store itself, which is LE for the GIC, as the
+ * guest knows the GIC is always LE.
+ *
+ * We convert this value to the CPUs native format to deal with it as a data
+ * value.
+ */
+unsigned long vgic_data_mmio_bus_to_host(const void *val, unsigned int len)
+{
+	unsigned long data = kvm_mmio_read_buf(val, len);
+
+	switch (len) {
+	case 1:
+		return data;
+	case 2:
+		return le16_to_cpu(data);
+	case 4:
+		return le32_to_cpu(data);
+	default:
+		return le64_to_cpu(data);
+	}
+}
+
+/*
+ * kvm_mmio_write_buf() expects a value in a format such that if converted to
+ * a byte array it is observed as the guest would see it if it could perform
+ * the load directly.  Since the GIC is LE, and the guest knows this, the
+ * guest expects a value in little endian format.
+ *
+ * We convert the data value from the CPUs native format to LE so that the
+ * value is returned in the proper format.
+ */
+void vgic_data_host_to_mmio_bus(void *buf, unsigned int len,
+				unsigned long data)
+{
+	switch (len) {
+	case 1:
+		break;
+	case 2:
+		data = cpu_to_le16(data);
+		break;
+	case 4:
+		data = cpu_to_le32(data);
+		break;
+	default:
+		data = cpu_to_le64(data);
+	}
+
+	kvm_mmio_write_buf(buf, len, data);
+}
+
+static
+struct vgic_io_device *kvm_to_vgic_iodev(const struct kvm_io_device *dev)
+{
+	return container_of(dev, struct vgic_io_device, dev);
+}
+
+static bool check_region(const struct kvm *kvm,
+			 const struct vgic_register_region *region,
+			 gpa_t addr, int len)
+{
+	int flags, nr_irqs = kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS;
+
+	switch (len) {
+	case sizeof(u8):
+		flags = VGIC_ACCESS_8bit;
+		break;
+	case sizeof(u32):
+		flags = VGIC_ACCESS_32bit;
+		break;
+	case sizeof(u64):
+		flags = VGIC_ACCESS_64bit;
+		break;
+	default:
+		return false;
+	}
+
+	if ((region->access_flags & flags) && IS_ALIGNED(addr, len)) {
+		if (!region->bits_per_irq)
+			return true;
+
+		/* Do we access a non-allocated IRQ? */
+		return VGIC_ADDR_TO_INTID(addr, region->bits_per_irq) < nr_irqs;
+	}
+
+	return false;
+}
+
+const struct vgic_register_region *
+vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev,
+		     gpa_t addr, int len)
+{
+	const struct vgic_register_region *region;
+
+	region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions,
+				       addr - iodev->base_addr);
+	if (!region || !check_region(vcpu->kvm, region, addr, len))
+		return NULL;
+
+	return region;
+}
+
+static int vgic_uaccess_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
+			     gpa_t addr, u32 *val)
+{
+	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
+	const struct vgic_register_region *region;
+	struct kvm_vcpu *r_vcpu;
+
+	region = vgic_get_mmio_region(vcpu, iodev, addr, sizeof(u32));
+	if (!region) {
+		*val = 0;
+		return 0;
+	}
+
+	r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu;
+	if (region->uaccess_read)
+		*val = region->uaccess_read(r_vcpu, addr, sizeof(u32));
+	else
+		*val = region->read(r_vcpu, addr, sizeof(u32));
+
+	return 0;
+}
+
+static int vgic_uaccess_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
+			      gpa_t addr, const u32 *val)
+{
+	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
+	const struct vgic_register_region *region;
+	struct kvm_vcpu *r_vcpu;
+
+	region = vgic_get_mmio_region(vcpu, iodev, addr, sizeof(u32));
+	if (!region)
+		return 0;
+
+	r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu;
+	if (region->uaccess_write)
+		return region->uaccess_write(r_vcpu, addr, sizeof(u32), *val);
+
+	region->write(r_vcpu, addr, sizeof(u32), *val);
+	return 0;
+}
+
+/*
+ * Userland access to VGIC registers.
+ */
+int vgic_uaccess(struct kvm_vcpu *vcpu, struct vgic_io_device *dev,
+		 bool is_write, int offset, u32 *val)
+{
+	if (is_write)
+		return vgic_uaccess_write(vcpu, &dev->dev, offset, val);
+	else
+		return vgic_uaccess_read(vcpu, &dev->dev, offset, val);
+}
+
+static int dispatch_mmio_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
+			      gpa_t addr, int len, void *val)
+{
+	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
+	const struct vgic_register_region *region;
+	unsigned long data = 0;
+
+	region = vgic_get_mmio_region(vcpu, iodev, addr, len);
+	if (!region) {
+		memset(val, 0, len);
+		return 0;
+	}
+
+	switch (iodev->iodev_type) {
+	case IODEV_CPUIF:
+		data = region->read(vcpu, addr, len);
+		break;
+	case IODEV_DIST:
+		data = region->read(vcpu, addr, len);
+		break;
+	case IODEV_REDIST:
+		data = region->read(iodev->redist_vcpu, addr, len);
+		break;
+	case IODEV_ITS:
+		data = region->its_read(vcpu->kvm, iodev->its, addr, len);
+		break;
+	}
+
+	vgic_data_host_to_mmio_bus(val, len, data);
+	return 0;
+}
+
+static int dispatch_mmio_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
+			       gpa_t addr, int len, const void *val)
+{
+	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
+	const struct vgic_register_region *region;
+	unsigned long data = vgic_data_mmio_bus_to_host(val, len);
+
+	region = vgic_get_mmio_region(vcpu, iodev, addr, len);
+	if (!region)
+		return 0;
+
+	switch (iodev->iodev_type) {
+	case IODEV_CPUIF:
+		region->write(vcpu, addr, len, data);
+		break;
+	case IODEV_DIST:
+		region->write(vcpu, addr, len, data);
+		break;
+	case IODEV_REDIST:
+		region->write(iodev->redist_vcpu, addr, len, data);
+		break;
+	case IODEV_ITS:
+		region->its_write(vcpu->kvm, iodev->its, addr, len, data);
+		break;
+	}
+
+	return 0;
+}
+
+struct kvm_io_device_ops kvm_io_gic_ops = {
+	.read = dispatch_mmio_read,
+	.write = dispatch_mmio_write,
+};
+
+int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
+			     enum vgic_type type)
+{
+	struct vgic_io_device *io_device = &kvm->arch.vgic.dist_iodev;
+	int ret = 0;
+	unsigned int len;
+
+	switch (type) {
+	case VGIC_V2:
+		len = vgic_v2_init_dist_iodev(io_device);
+		break;
+	case VGIC_V3:
+		len = vgic_v3_init_dist_iodev(io_device);
+		break;
+	default:
+		BUG_ON(1);
+	}
+
+	io_device->base_addr = dist_base_address;
+	io_device->iodev_type = IODEV_DIST;
+	io_device->redist_vcpu = NULL;
+
+	mutex_lock(&kvm->slots_lock);
+	ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, dist_base_address,
+				      len, &io_device->dev);
+	mutex_unlock(&kvm->slots_lock);
+
+	return ret;
+}

virt/kvm/arm/vgic/vgic-mmio.h

@@ -0,0 +1,216 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ */
+#ifndef __KVM_ARM_VGIC_MMIO_H__
+#define __KVM_ARM_VGIC_MMIO_H__
+
+struct vgic_register_region {
+	unsigned int reg_offset;
+	unsigned int len;
+	unsigned int bits_per_irq;
+	unsigned int access_flags;
+	union {
+		unsigned long (*read)(struct kvm_vcpu *vcpu, gpa_t addr,
+				      unsigned int len);
+		unsigned long (*its_read)(struct kvm *kvm, struct vgic_its *its,
+					  gpa_t addr, unsigned int len);
+	};
+	union {
+		void (*write)(struct kvm_vcpu *vcpu, gpa_t addr,
+			      unsigned int len, unsigned long val);
+		void (*its_write)(struct kvm *kvm, struct vgic_its *its,
+				  gpa_t addr, unsigned int len,
+				  unsigned long val);
+	};
+	unsigned long (*uaccess_read)(struct kvm_vcpu *vcpu, gpa_t addr,
+				      unsigned int len);
+	union {
+		int (*uaccess_write)(struct kvm_vcpu *vcpu, gpa_t addr,
+				     unsigned int len, unsigned long val);
+		int (*uaccess_its_write)(struct kvm *kvm, struct vgic_its *its,
+					 gpa_t addr, unsigned int len,
+					 unsigned long val);
+	};
+};
+
+extern struct kvm_io_device_ops kvm_io_gic_ops;
+
+#define VGIC_ACCESS_8bit	1
+#define VGIC_ACCESS_32bit	2
+#define VGIC_ACCESS_64bit	4
+
+/*
+ * Generate a mask that covers the number of bytes required to address
+ * up to 1024 interrupts, each represented by <bits> bits. This assumes
+ * that <bits> is a power of two.
+ */
+#define VGIC_ADDR_IRQ_MASK(bits) (((bits) * 1024 / 8) - 1)
+
+/*
+ * (addr & mask) gives us the _byte_ offset for the INT ID.
+ * We multiply this by 8 the get the _bit_ offset, then divide this by
+ * the number of bits to learn the actual INT ID.
+ * But instead of a division (which requires a "long long div" implementation),
+ * we shift by the binary logarithm of <bits>.
+ * This assumes that <bits> is a power of two.
+ */
+#define VGIC_ADDR_TO_INTID(addr, bits)  (((addr) & VGIC_ADDR_IRQ_MASK(bits)) * \
+					8 >> ilog2(bits))
+
+/*
+ * Some VGIC registers store per-IRQ information, with a different number
+ * of bits per IRQ. For those registers this macro is used.
+ * The _WITH_LENGTH version instantiates registers with a fixed length
+ * and is mutually exclusive with the _PER_IRQ version.
+ */
+#define REGISTER_DESC_WITH_BITS_PER_IRQ(off, rd, wr, ur, uw, bpi, acc)	\
+	{								\
+		.reg_offset = off,					\
+		.bits_per_irq = bpi,					\
+		.len = bpi * 1024 / 8,					\
+		.access_flags = acc,					\
+		.read = rd,						\
+		.write = wr,						\
+		.uaccess_read = ur,					\
+		.uaccess_write = uw,					\
+	}
+
+#define REGISTER_DESC_WITH_LENGTH(off, rd, wr, length, acc)		\
+	{								\
+		.reg_offset = off,					\
+		.bits_per_irq = 0,					\
+		.len = length,						\
+		.access_flags = acc,					\
+		.read = rd,						\
+		.write = wr,						\
+	}
+
+#define REGISTER_DESC_WITH_LENGTH_UACCESS(off, rd, wr, urd, uwr, length, acc) \
+	{								\
+		.reg_offset = off,					\
+		.bits_per_irq = 0,					\
+		.len = length,						\
+		.access_flags = acc,					\
+		.read = rd,						\
+		.write = wr,						\
+		.uaccess_read = urd,					\
+		.uaccess_write = uwr,					\
+	}
+
+int kvm_vgic_register_mmio_region(struct kvm *kvm, struct kvm_vcpu *vcpu,
+				  struct vgic_register_region *reg_desc,
+				  struct vgic_io_device *region,
+				  int nr_irqs, bool offset_private);
+
+unsigned long vgic_data_mmio_bus_to_host(const void *val, unsigned int len);
+
+void vgic_data_host_to_mmio_bus(void *buf, unsigned int len,
+				unsigned long data);
+
+unsigned long extract_bytes(u64 data, unsigned int offset,
+			    unsigned int num);
+
+u64 update_64bit_reg(u64 reg, unsigned int offset, unsigned int len,
+		     unsigned long val);
+
+unsigned long vgic_mmio_read_raz(struct kvm_vcpu *vcpu,
+				 gpa_t addr, unsigned int len);
+
+unsigned long vgic_mmio_read_rao(struct kvm_vcpu *vcpu,
+				 gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_wi(struct kvm_vcpu *vcpu, gpa_t addr,
+			unsigned int len, unsigned long val);
+
+int vgic_mmio_uaccess_write_wi(struct kvm_vcpu *vcpu, gpa_t addr,
+			       unsigned int len, unsigned long val);
+
+unsigned long vgic_mmio_read_group(struct kvm_vcpu *vcpu, gpa_t addr,
+				   unsigned int len);
+
+void vgic_mmio_write_group(struct kvm_vcpu *vcpu, gpa_t addr,
+			   unsigned int len, unsigned long val);
+
+unsigned long vgic_mmio_read_enable(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_senable(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val);
+
+void vgic_mmio_write_cenable(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val);
+
+unsigned long vgic_mmio_read_pending(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_spending(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val);
+
+void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val);
+
+unsigned long vgic_mmio_read_active(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len);
+
+unsigned long vgic_uaccess_read_active(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_cactive(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val);
+
+void vgic_mmio_write_sactive(struct kvm_vcpu *vcpu,
+			     gpa_t addr, unsigned int len,
+			     unsigned long val);
+
+int vgic_mmio_uaccess_write_cactive(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len,
+				    unsigned long val);
+
+int vgic_mmio_uaccess_write_sactive(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len,
+				    unsigned long val);
+
+unsigned long vgic_mmio_read_priority(struct kvm_vcpu *vcpu,
+				      gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_priority(struct kvm_vcpu *vcpu,
+			      gpa_t addr, unsigned int len,
+			      unsigned long val);
+
+unsigned long vgic_mmio_read_config(struct kvm_vcpu *vcpu,
+				    gpa_t addr, unsigned int len);
+
+void vgic_mmio_write_config(struct kvm_vcpu *vcpu,
+			    gpa_t addr, unsigned int len,
+			    unsigned long val);
+
+int vgic_uaccess(struct kvm_vcpu *vcpu, struct vgic_io_device *dev,
+		 bool is_write, int offset, u32 *val);
+
+u64 vgic_read_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid);
+
+void vgic_write_irq_line_level_info(struct kvm_vcpu *vcpu, u32 intid,
+				    const u64 val);
+
+unsigned int vgic_v2_init_dist_iodev(struct vgic_io_device *dev);
+
+unsigned int vgic_v3_init_dist_iodev(struct vgic_io_device *dev);
+
+u64 vgic_sanitise_outer_cacheability(u64 reg);
+u64 vgic_sanitise_inner_cacheability(u64 reg);
+u64 vgic_sanitise_shareability(u64 reg);
+u64 vgic_sanitise_field(u64 reg, u64 field_mask, int field_shift,
+			u64 (*sanitise_fn)(u64));
+
+/* Find the proper register handler entry given a certain address offset */
+const struct vgic_register_region *
+vgic_find_mmio_region(const struct vgic_register_region *regions,
+		      int nr_regions, unsigned int offset);
+
+#endif

virt/kvm/arm/vgic/vgic-v2.c

@@ -0,0 +1,504 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ */
+
+#include <linux/irqchip/arm-gic.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <kvm/arm_vgic.h>
+#include <asm/kvm_mmu.h>
+
+#include "vgic.h"
+
+static inline void vgic_v2_write_lr(int lr, u32 val)
+{
+	void __iomem *base = kvm_vgic_global_state.vctrl_base;
+
+	writel_relaxed(val, base + GICH_LR0 + (lr * 4));
+}
+
+void vgic_v2_init_lrs(void)
+{
+	int i;
+
+	for (i = 0; i < kvm_vgic_global_state.nr_lr; i++)
+		vgic_v2_write_lr(i, 0);
+}
+
+void vgic_v2_set_underflow(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v2;
+
+	cpuif->vgic_hcr |= GICH_HCR_UIE;
+}
+
+static bool lr_signals_eoi_mi(u32 lr_val)
+{
+	return !(lr_val & GICH_LR_STATE) && (lr_val & GICH_LR_EOI) &&
+	       !(lr_val & GICH_LR_HW);
+}
+
+/*
+ * transfer the content of the LRs back into the corresponding ap_list:
+ * - active bit is transferred as is
+ * - pending bit is
+ *   - transferred as is in case of edge sensitive IRQs
+ *   - set to the line-level (resample time) for level sensitive IRQs
+ */
+void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_v2_cpu_if *cpuif = &vgic_cpu->vgic_v2;
+	int lr;
+
+	DEBUG_SPINLOCK_BUG_ON(!irqs_disabled());
+
+	cpuif->vgic_hcr &= ~GICH_HCR_UIE;
+
+	for (lr = 0; lr < vgic_cpu->used_lrs; lr++) {
+		u32 val = cpuif->vgic_lr[lr];
+		u32 cpuid, intid = val & GICH_LR_VIRTUALID;
+		struct vgic_irq *irq;
+
+		/* Extract the source vCPU id from the LR */
+		cpuid = val & GICH_LR_PHYSID_CPUID;
+		cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT;
+		cpuid &= 7;
+
+		/* Notify fds when the guest EOI'ed a level-triggered SPI */
+		if (lr_signals_eoi_mi(val) && vgic_valid_spi(vcpu->kvm, intid))
+			kvm_notify_acked_irq(vcpu->kvm, 0,
+					     intid - VGIC_NR_PRIVATE_IRQS);
+
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid);
+
+		raw_spin_lock(&irq->irq_lock);
+
+		/* Always preserve the active bit */
+		irq->active = !!(val & GICH_LR_ACTIVE_BIT);
+
+		if (irq->active && vgic_irq_is_sgi(intid))
+			irq->active_source = cpuid;
+
+		/* Edge is the only case where we preserve the pending bit */
+		if (irq->config == VGIC_CONFIG_EDGE &&
+		    (val & GICH_LR_PENDING_BIT)) {
+			irq->pending_latch = true;
+
+			if (vgic_irq_is_sgi(intid))
+				irq->source |= (1 << cpuid);
+		}
+
+		/*
+		 * Clear soft pending state when level irqs have been acked.
+		 */
+		if (irq->config == VGIC_CONFIG_LEVEL && !(val & GICH_LR_STATE))
+			irq->pending_latch = false;
+
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+
+		raw_spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	vgic_cpu->used_lrs = 0;
+}
+
+/*
+ * Populates the particular LR with the state of a given IRQ:
+ * - for an edge sensitive IRQ the pending state is cleared in struct vgic_irq
+ * - for a level sensitive IRQ the pending state value is unchanged;
+ *   it is dictated directly by the input level
+ *
+ * If @irq describes an SGI with multiple sources, we choose the
+ * lowest-numbered source VCPU and clear that bit in the source bitmap.
+ *
+ * The irq_lock must be held by the caller.
+ */
+void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
+{
+	u32 val = irq->intid;
+	bool allow_pending = true;
+
+	if (irq->active) {
+		val |= GICH_LR_ACTIVE_BIT;
+		if (vgic_irq_is_sgi(irq->intid))
+			val |= irq->active_source << GICH_LR_PHYSID_CPUID_SHIFT;
+		if (vgic_irq_is_multi_sgi(irq)) {
+			allow_pending = false;
+			val |= GICH_LR_EOI;
+		}
+	}
+
+	if (irq->group)
+		val |= GICH_LR_GROUP1;
+
+	if (irq->hw) {
+		val |= GICH_LR_HW;
+		val |= irq->hwintid << GICH_LR_PHYSID_CPUID_SHIFT;
+		/*
+		 * Never set pending+active on a HW interrupt, as the
+		 * pending state is kept at the physical distributor
+		 * level.
+		 */
+		if (irq->active)
+			allow_pending = false;
+	} else {
+		if (irq->config == VGIC_CONFIG_LEVEL) {
+			val |= GICH_LR_EOI;
+
+			/*
+			 * Software resampling doesn't work very well
+			 * if we allow P+A, so let's not do that.
+			 */
+			if (irq->active)
+				allow_pending = false;
+		}
+	}
+
+	if (allow_pending && irq_is_pending(irq)) {
+		val |= GICH_LR_PENDING_BIT;
+
+		if (irq->config == VGIC_CONFIG_EDGE)
+			irq->pending_latch = false;
+
+		if (vgic_irq_is_sgi(irq->intid)) {
+			u32 src = ffs(irq->source);
+
+			if (WARN_RATELIMIT(!src, "No SGI source for INTID %d\n",
+					   irq->intid))
+				return;
+
+			val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT;
+			irq->source &= ~(1 << (src - 1));
+			if (irq->source) {
+				irq->pending_latch = true;
+				val |= GICH_LR_EOI;
+			}
+		}
+	}
+
+	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v2_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
+	/* The GICv2 LR only holds five bits of priority. */
+	val |= (irq->priority >> 3) << GICH_LR_PRIORITY_SHIFT;
+
+	vcpu->arch.vgic_cpu.vgic_v2.vgic_lr[lr] = val;
+}
+
+void vgic_v2_clear_lr(struct kvm_vcpu *vcpu, int lr)
+{
+	vcpu->arch.vgic_cpu.vgic_v2.vgic_lr[lr] = 0;
+}
+
+void vgic_v2_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+	u32 vmcr;
+
+	vmcr = (vmcrp->grpen0 << GICH_VMCR_ENABLE_GRP0_SHIFT) &
+		GICH_VMCR_ENABLE_GRP0_MASK;
+	vmcr |= (vmcrp->grpen1 << GICH_VMCR_ENABLE_GRP1_SHIFT) &
+		GICH_VMCR_ENABLE_GRP1_MASK;
+	vmcr |= (vmcrp->ackctl << GICH_VMCR_ACK_CTL_SHIFT) &
+		GICH_VMCR_ACK_CTL_MASK;
+	vmcr |= (vmcrp->fiqen << GICH_VMCR_FIQ_EN_SHIFT) &
+		GICH_VMCR_FIQ_EN_MASK;
+	vmcr |= (vmcrp->cbpr << GICH_VMCR_CBPR_SHIFT) &
+		GICH_VMCR_CBPR_MASK;
+	vmcr |= (vmcrp->eoim << GICH_VMCR_EOI_MODE_SHIFT) &
+		GICH_VMCR_EOI_MODE_MASK;
+	vmcr |= (vmcrp->abpr << GICH_VMCR_ALIAS_BINPOINT_SHIFT) &
+		GICH_VMCR_ALIAS_BINPOINT_MASK;
+	vmcr |= (vmcrp->bpr << GICH_VMCR_BINPOINT_SHIFT) &
+		GICH_VMCR_BINPOINT_MASK;
+	vmcr |= ((vmcrp->pmr >> GICV_PMR_PRIORITY_SHIFT) <<
+		 GICH_VMCR_PRIMASK_SHIFT) & GICH_VMCR_PRIMASK_MASK;
+
+	cpu_if->vgic_vmcr = vmcr;
+}
+
+void vgic_v2_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+	u32 vmcr;
+
+	vmcr = cpu_if->vgic_vmcr;
+
+	vmcrp->grpen0 = (vmcr & GICH_VMCR_ENABLE_GRP0_MASK) >>
+		GICH_VMCR_ENABLE_GRP0_SHIFT;
+	vmcrp->grpen1 = (vmcr & GICH_VMCR_ENABLE_GRP1_MASK) >>
+		GICH_VMCR_ENABLE_GRP1_SHIFT;
+	vmcrp->ackctl = (vmcr & GICH_VMCR_ACK_CTL_MASK) >>
+		GICH_VMCR_ACK_CTL_SHIFT;
+	vmcrp->fiqen = (vmcr & GICH_VMCR_FIQ_EN_MASK) >>
+		GICH_VMCR_FIQ_EN_SHIFT;
+	vmcrp->cbpr = (vmcr & GICH_VMCR_CBPR_MASK) >>
+		GICH_VMCR_CBPR_SHIFT;
+	vmcrp->eoim = (vmcr & GICH_VMCR_EOI_MODE_MASK) >>
+		GICH_VMCR_EOI_MODE_SHIFT;
+
+	vmcrp->abpr = (vmcr & GICH_VMCR_ALIAS_BINPOINT_MASK) >>
+			GICH_VMCR_ALIAS_BINPOINT_SHIFT;
+	vmcrp->bpr  = (vmcr & GICH_VMCR_BINPOINT_MASK) >>
+			GICH_VMCR_BINPOINT_SHIFT;
+	vmcrp->pmr  = ((vmcr & GICH_VMCR_PRIMASK_MASK) >>
+			GICH_VMCR_PRIMASK_SHIFT) << GICV_PMR_PRIORITY_SHIFT;
+}
+
+void vgic_v2_enable(struct kvm_vcpu *vcpu)
+{
+	/*
+	 * By forcing VMCR to zero, the GIC will restore the binary
+	 * points to their reset values. Anything else resets to zero
+	 * anyway.
+	 */
+	vcpu->arch.vgic_cpu.vgic_v2.vgic_vmcr = 0;
+
+	/* Get the show on the road... */
+	vcpu->arch.vgic_cpu.vgic_v2.vgic_hcr = GICH_HCR_EN;
+}
+
+/* check for overlapping regions and for regions crossing the end of memory */
+static bool vgic_v2_check_base(gpa_t dist_base, gpa_t cpu_base)
+{
+	if (dist_base + KVM_VGIC_V2_DIST_SIZE < dist_base)
+		return false;
+	if (cpu_base + KVM_VGIC_V2_CPU_SIZE < cpu_base)
+		return false;
+
+	if (dist_base + KVM_VGIC_V2_DIST_SIZE <= cpu_base)
+		return true;
+	if (cpu_base + KVM_VGIC_V2_CPU_SIZE <= dist_base)
+		return true;
+
+	return false;
+}
+
+int vgic_v2_map_resources(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	int ret = 0;
+
+	if (vgic_ready(kvm))
+		goto out;
+
+	if (IS_VGIC_ADDR_UNDEF(dist->vgic_dist_base) ||
+	    IS_VGIC_ADDR_UNDEF(dist->vgic_cpu_base)) {
+		kvm_err("Need to set vgic cpu and dist addresses first\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	if (!vgic_v2_check_base(dist->vgic_dist_base, dist->vgic_cpu_base)) {
+		kvm_err("VGIC CPU and dist frames overlap\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * Initialize the vgic if this hasn't already been done on demand by
+	 * accessing the vgic state from userspace.
+	 */
+	ret = vgic_init(kvm);
+	if (ret) {
+		kvm_err("Unable to initialize VGIC dynamic data structures\n");
+		goto out;
+	}
+
+	ret = vgic_register_dist_iodev(kvm, dist->vgic_dist_base, VGIC_V2);
+	if (ret) {
+		kvm_err("Unable to register VGIC MMIO regions\n");
+		goto out;
+	}
+
+	if (!static_branch_unlikely(&vgic_v2_cpuif_trap)) {
+		ret = kvm_phys_addr_ioremap(kvm, dist->vgic_cpu_base,
+					    kvm_vgic_global_state.vcpu_base,
+					    KVM_VGIC_V2_CPU_SIZE, true);
+		if (ret) {
+			kvm_err("Unable to remap VGIC CPU to VCPU\n");
+			goto out;
+		}
+	}
+
+	dist->ready = true;
+
+out:
+	return ret;
+}
+
+DEFINE_STATIC_KEY_FALSE(vgic_v2_cpuif_trap);
+
+/**
+ * vgic_v2_probe - probe for a VGICv2 compatible interrupt controller
+ * @info:	pointer to the GIC description
+ *
+ * Returns 0 if the VGICv2 has been probed successfully, returns an error code
+ * otherwise
+ */
+int vgic_v2_probe(const struct gic_kvm_info *info)
+{
+	int ret;
+	u32 vtr;
+
+	if (!info->vctrl.start) {
+		kvm_err("GICH not present in the firmware table\n");
+		return -ENXIO;
+	}
+
+	if (!PAGE_ALIGNED(info->vcpu.start) ||
+	    !PAGE_ALIGNED(resource_size(&info->vcpu))) {
+		kvm_info("GICV region size/alignment is unsafe, using trapping (reduced performance)\n");
+
+		ret = create_hyp_io_mappings(info->vcpu.start,
+					     resource_size(&info->vcpu),
+					     &kvm_vgic_global_state.vcpu_base_va,
+					     &kvm_vgic_global_state.vcpu_hyp_va);
+		if (ret) {
+			kvm_err("Cannot map GICV into hyp\n");
+			goto out;
+		}
+
+		static_branch_enable(&vgic_v2_cpuif_trap);
+	}
+
+	ret = create_hyp_io_mappings(info->vctrl.start,
+				     resource_size(&info->vctrl),
+				     &kvm_vgic_global_state.vctrl_base,
+				     &kvm_vgic_global_state.vctrl_hyp);
+	if (ret) {
+		kvm_err("Cannot map VCTRL into hyp\n");
+		goto out;
+	}
+
+	vtr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_VTR);
+	kvm_vgic_global_state.nr_lr = (vtr & 0x3f) + 1;
+
+	ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+	if (ret) {
+		kvm_err("Cannot register GICv2 KVM device\n");
+		goto out;
+	}
+
+	kvm_vgic_global_state.can_emulate_gicv2 = true;
+	kvm_vgic_global_state.vcpu_base = info->vcpu.start;
+	kvm_vgic_global_state.type = VGIC_V2;
+	kvm_vgic_global_state.max_gic_vcpus = VGIC_V2_MAX_CPUS;
+
+	kvm_debug("vgic-v2@%llx\n", info->vctrl.start);
+
+	return 0;
+out:
+	if (kvm_vgic_global_state.vctrl_base)
+		iounmap(kvm_vgic_global_state.vctrl_base);
+	if (kvm_vgic_global_state.vcpu_base_va)
+		iounmap(kvm_vgic_global_state.vcpu_base_va);
+
+	return ret;
+}
+
+static void save_lrs(struct kvm_vcpu *vcpu, void __iomem *base)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	u64 elrsr;
+	int i;
+
+	elrsr = readl_relaxed(base + GICH_ELRSR0);
+	if (unlikely(used_lrs > 32))
+		elrsr |= ((u64)readl_relaxed(base + GICH_ELRSR1)) << 32;
+
+	for (i = 0; i < used_lrs; i++) {
+		if (elrsr & (1UL << i))
+			cpu_if->vgic_lr[i] &= ~GICH_LR_STATE;
+		else
+			cpu_if->vgic_lr[i] = readl_relaxed(base + GICH_LR0 + (i * 4));
+
+		writel_relaxed(0, base + GICH_LR0 + (i * 4));
+	}
+}
+
+void vgic_v2_save_state(struct kvm_vcpu *vcpu)
+{
+	void __iomem *base = kvm_vgic_global_state.vctrl_base;
+	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+
+	if (!base)
+		return;
+
+	if (used_lrs) {
+		save_lrs(vcpu, base);
+		writel_relaxed(0, base + GICH_HCR);
+	}
+}
+
+void vgic_v2_restore_state(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+	void __iomem *base = kvm_vgic_global_state.vctrl_base;
+	u64 used_lrs = vcpu->arch.vgic_cpu.used_lrs;
+	int i;
+
+	if (!base)
+		return;
+
+	if (used_lrs) {
+		writel_relaxed(cpu_if->vgic_hcr, base + GICH_HCR);
+		for (i = 0; i < used_lrs; i++) {
+			writel_relaxed(cpu_if->vgic_lr[i],
+				       base + GICH_LR0 + (i * 4));
+		}
+	}
+}
+
+void vgic_v2_load(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+
+	writel_relaxed(cpu_if->vgic_vmcr,
+		       kvm_vgic_global_state.vctrl_base + GICH_VMCR);
+	writel_relaxed(cpu_if->vgic_apr,
+		       kvm_vgic_global_state.vctrl_base + GICH_APR);
+}
+
+void vgic_v2_vmcr_sync(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+
+	cpu_if->vgic_vmcr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_VMCR);
+}
+
+void vgic_v2_put(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+
+	vgic_v2_vmcr_sync(vcpu);
+	cpu_if->vgic_apr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_APR);
+}

virt/kvm/arm/vgic/vgic-v3.c

@@ -0,0 +1,685 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/irqchip/arm-gic-v3.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <kvm/arm_vgic.h>
+#include <asm/kvm_hyp.h>
+#include <asm/kvm_mmu.h>
+#include <asm/kvm_asm.h>
+
+#include "vgic.h"
+
+static bool group0_trap;
+static bool group1_trap;
+static bool common_trap;
+static bool gicv4_enable;
+
+void vgic_v3_set_underflow(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v3_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v3;
+
+	cpuif->vgic_hcr |= ICH_HCR_UIE;
+}
+
+static bool lr_signals_eoi_mi(u64 lr_val)
+{
+	return !(lr_val & ICH_LR_STATE) && (lr_val & ICH_LR_EOI) &&
+	       !(lr_val & ICH_LR_HW);
+}
+
+void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	struct vgic_v3_cpu_if *cpuif = &vgic_cpu->vgic_v3;
+	u32 model = vcpu->kvm->arch.vgic.vgic_model;
+	int lr;
+
+	DEBUG_SPINLOCK_BUG_ON(!irqs_disabled());
+
+	cpuif->vgic_hcr &= ~ICH_HCR_UIE;
+
+	for (lr = 0; lr < vgic_cpu->used_lrs; lr++) {
+		u64 val = cpuif->vgic_lr[lr];
+		u32 intid, cpuid;
+		struct vgic_irq *irq;
+		bool is_v2_sgi = false;
+
+		cpuid = val & GICH_LR_PHYSID_CPUID;
+		cpuid >>= GICH_LR_PHYSID_CPUID_SHIFT;
+
+		if (model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+			intid = val & ICH_LR_VIRTUAL_ID_MASK;
+		} else {
+			intid = val & GICH_LR_VIRTUALID;
+			is_v2_sgi = vgic_irq_is_sgi(intid);
+		}
+
+		/* Notify fds when the guest EOI'ed a level-triggered IRQ */
+		if (lr_signals_eoi_mi(val) && vgic_valid_spi(vcpu->kvm, intid))
+			kvm_notify_acked_irq(vcpu->kvm, 0,
+					     intid - VGIC_NR_PRIVATE_IRQS);
+
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid);
+		if (!irq)	/* An LPI could have been unmapped. */
+			continue;
+
+		raw_spin_lock(&irq->irq_lock);
+
+		/* Always preserve the active bit */
+		irq->active = !!(val & ICH_LR_ACTIVE_BIT);
+
+		if (irq->active && is_v2_sgi)
+			irq->active_source = cpuid;
+
+		/* Edge is the only case where we preserve the pending bit */
+		if (irq->config == VGIC_CONFIG_EDGE &&
+		    (val & ICH_LR_PENDING_BIT)) {
+			irq->pending_latch = true;
+
+			if (is_v2_sgi)
+				irq->source |= (1 << cpuid);
+		}
+
+		/*
+		 * Clear soft pending state when level irqs have been acked.
+		 */
+		if (irq->config == VGIC_CONFIG_LEVEL && !(val & ICH_LR_STATE))
+			irq->pending_latch = false;
+
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+
+		raw_spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	vgic_cpu->used_lrs = 0;
+}
+
+/* Requires the irq to be locked already */
+void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
+{
+	u32 model = vcpu->kvm->arch.vgic.vgic_model;
+	u64 val = irq->intid;
+	bool allow_pending = true, is_v2_sgi;
+
+	is_v2_sgi = (vgic_irq_is_sgi(irq->intid) &&
+		     model == KVM_DEV_TYPE_ARM_VGIC_V2);
+
+	if (irq->active) {
+		val |= ICH_LR_ACTIVE_BIT;
+		if (is_v2_sgi)
+			val |= irq->active_source << GICH_LR_PHYSID_CPUID_SHIFT;
+		if (vgic_irq_is_multi_sgi(irq)) {
+			allow_pending = false;
+			val |= ICH_LR_EOI;
+		}
+	}
+
+	if (irq->hw) {
+		val |= ICH_LR_HW;
+		val |= ((u64)irq->hwintid) << ICH_LR_PHYS_ID_SHIFT;
+		/*
+		 * Never set pending+active on a HW interrupt, as the
+		 * pending state is kept at the physical distributor
+		 * level.
+		 */
+		if (irq->active)
+			allow_pending = false;
+	} else {
+		if (irq->config == VGIC_CONFIG_LEVEL) {
+			val |= ICH_LR_EOI;
+
+			/*
+			 * Software resampling doesn't work very well
+			 * if we allow P+A, so let's not do that.
+			 */
+			if (irq->active)
+				allow_pending = false;
+		}
+	}
+
+	if (allow_pending && irq_is_pending(irq)) {
+		val |= ICH_LR_PENDING_BIT;
+
+		if (irq->config == VGIC_CONFIG_EDGE)
+			irq->pending_latch = false;
+
+		if (vgic_irq_is_sgi(irq->intid) &&
+		    model == KVM_DEV_TYPE_ARM_VGIC_V2) {
+			u32 src = ffs(irq->source);
+
+			if (WARN_RATELIMIT(!src, "No SGI source for INTID %d\n",
+					   irq->intid))
+				return;
+
+			val |= (src - 1) << GICH_LR_PHYSID_CPUID_SHIFT;
+			irq->source &= ~(1 << (src - 1));
+			if (irq->source) {
+				irq->pending_latch = true;
+				val |= ICH_LR_EOI;
+			}
+		}
+	}
+
+	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v3_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
+	if (irq->group)
+		val |= ICH_LR_GROUP;
+
+	val |= (u64)irq->priority << ICH_LR_PRIORITY_SHIFT;
+
+	vcpu->arch.vgic_cpu.vgic_v3.vgic_lr[lr] = val;
+}
+
+void vgic_v3_clear_lr(struct kvm_vcpu *vcpu, int lr)
+{
+	vcpu->arch.vgic_cpu.vgic_v3.vgic_lr[lr] = 0;
+}
+
+void vgic_v3_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
+{
+	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+	u32 model = vcpu->kvm->arch.vgic.vgic_model;
+	u32 vmcr;
+
+	if (model == KVM_DEV_TYPE_ARM_VGIC_V2) {
+		vmcr = (vmcrp->ackctl << ICH_VMCR_ACK_CTL_SHIFT) &
+			ICH_VMCR_ACK_CTL_MASK;
+		vmcr |= (vmcrp->fiqen << ICH_VMCR_FIQ_EN_SHIFT) &
+			ICH_VMCR_FIQ_EN_MASK;
+	} else {
+		/*
+		 * When emulating GICv3 on GICv3 with SRE=1 on the
+		 * VFIQEn bit is RES1 and the VAckCtl bit is RES0.
+		 */
+		vmcr = ICH_VMCR_FIQ_EN_MASK;
+	}
+
+	vmcr |= (vmcrp->cbpr << ICH_VMCR_CBPR_SHIFT) & ICH_VMCR_CBPR_MASK;
+	vmcr |= (vmcrp->eoim << ICH_VMCR_EOIM_SHIFT) & ICH_VMCR_EOIM_MASK;
+	vmcr |= (vmcrp->abpr << ICH_VMCR_BPR1_SHIFT) & ICH_VMCR_BPR1_MASK;
+	vmcr |= (vmcrp->bpr << ICH_VMCR_BPR0_SHIFT) & ICH_VMCR_BPR0_MASK;
+	vmcr |= (vmcrp->pmr << ICH_VMCR_PMR_SHIFT) & ICH_VMCR_PMR_MASK;
+	vmcr |= (vmcrp->grpen0 << ICH_VMCR_ENG0_SHIFT) & ICH_VMCR_ENG0_MASK;
+	vmcr |= (vmcrp->grpen1 << ICH_VMCR_ENG1_SHIFT) & ICH_VMCR_ENG1_MASK;
+
+	cpu_if->vgic_vmcr = vmcr;
+}
+
+void vgic_v3_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
+{
+	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+	u32 model = vcpu->kvm->arch.vgic.vgic_model;
+	u32 vmcr;
+
+	vmcr = cpu_if->vgic_vmcr;
+
+	if (model == KVM_DEV_TYPE_ARM_VGIC_V2) {
+		vmcrp->ackctl = (vmcr & ICH_VMCR_ACK_CTL_MASK) >>
+			ICH_VMCR_ACK_CTL_SHIFT;
+		vmcrp->fiqen = (vmcr & ICH_VMCR_FIQ_EN_MASK) >>
+			ICH_VMCR_FIQ_EN_SHIFT;
+	} else {
+		/*
+		 * When emulating GICv3 on GICv3 with SRE=1 on the
+		 * VFIQEn bit is RES1 and the VAckCtl bit is RES0.
+		 */
+		vmcrp->fiqen = 1;
+		vmcrp->ackctl = 0;
+	}
+
+	vmcrp->cbpr = (vmcr & ICH_VMCR_CBPR_MASK) >> ICH_VMCR_CBPR_SHIFT;
+	vmcrp->eoim = (vmcr & ICH_VMCR_EOIM_MASK) >> ICH_VMCR_EOIM_SHIFT;
+	vmcrp->abpr = (vmcr & ICH_VMCR_BPR1_MASK) >> ICH_VMCR_BPR1_SHIFT;
+	vmcrp->bpr  = (vmcr & ICH_VMCR_BPR0_MASK) >> ICH_VMCR_BPR0_SHIFT;
+	vmcrp->pmr  = (vmcr & ICH_VMCR_PMR_MASK) >> ICH_VMCR_PMR_SHIFT;
+	vmcrp->grpen0 = (vmcr & ICH_VMCR_ENG0_MASK) >> ICH_VMCR_ENG0_SHIFT;
+	vmcrp->grpen1 = (vmcr & ICH_VMCR_ENG1_MASK) >> ICH_VMCR_ENG1_SHIFT;
+}
+
+#define INITIAL_PENDBASER_VALUE						  \
+	(GIC_BASER_CACHEABILITY(GICR_PENDBASER, INNER, RaWb)		| \
+	GIC_BASER_CACHEABILITY(GICR_PENDBASER, OUTER, SameAsInner)	| \
+	GIC_BASER_SHAREABILITY(GICR_PENDBASER, InnerShareable))
+
+void vgic_v3_enable(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v3_cpu_if *vgic_v3 = &vcpu->arch.vgic_cpu.vgic_v3;
+
+	/*
+	 * By forcing VMCR to zero, the GIC will restore the binary
+	 * points to their reset values. Anything else resets to zero
+	 * anyway.
+	 */
+	vgic_v3->vgic_vmcr = 0;
+
+	/*
+	 * If we are emulating a GICv3, we do it in an non-GICv2-compatible
+	 * way, so we force SRE to 1 to demonstrate this to the guest.
+	 * Also, we don't support any form of IRQ/FIQ bypass.
+	 * This goes with the spec allowing the value to be RAO/WI.
+	 */
+	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
+		vgic_v3->vgic_sre = (ICC_SRE_EL1_DIB |
+				     ICC_SRE_EL1_DFB |
+				     ICC_SRE_EL1_SRE);
+		vcpu->arch.vgic_cpu.pendbaser = INITIAL_PENDBASER_VALUE;
+	} else {
+		vgic_v3->vgic_sre = 0;
+	}
+
+	vcpu->arch.vgic_cpu.num_id_bits = (kvm_vgic_global_state.ich_vtr_el2 &
+					   ICH_VTR_ID_BITS_MASK) >>
+					   ICH_VTR_ID_BITS_SHIFT;
+	vcpu->arch.vgic_cpu.num_pri_bits = ((kvm_vgic_global_state.ich_vtr_el2 &
+					    ICH_VTR_PRI_BITS_MASK) >>
+					    ICH_VTR_PRI_BITS_SHIFT) + 1;
+
+	/* Get the show on the road... */
+	vgic_v3->vgic_hcr = ICH_HCR_EN;
+	if (group0_trap)
+		vgic_v3->vgic_hcr |= ICH_HCR_TALL0;
+	if (group1_trap)
+		vgic_v3->vgic_hcr |= ICH_HCR_TALL1;
+	if (common_trap)
+		vgic_v3->vgic_hcr |= ICH_HCR_TC;
+}
+
+int vgic_v3_lpi_sync_pending_status(struct kvm *kvm, struct vgic_irq *irq)
+{
+	struct kvm_vcpu *vcpu;
+	int byte_offset, bit_nr;
+	gpa_t pendbase, ptr;
+	bool status;
+	u8 val;
+	int ret;
+	unsigned long flags;
+
+retry:
+	vcpu = irq->target_vcpu;
+	if (!vcpu)
+		return 0;
+
+	pendbase = GICR_PENDBASER_ADDRESS(vcpu->arch.vgic_cpu.pendbaser);
+
+	byte_offset = irq->intid / BITS_PER_BYTE;
+	bit_nr = irq->intid % BITS_PER_BYTE;
+	ptr = pendbase + byte_offset;
+
+	ret = kvm_read_guest_lock(kvm, ptr, &val, 1);
+	if (ret)
+		return ret;
+
+	status = val & (1 << bit_nr);
+
+	raw_spin_lock_irqsave(&irq->irq_lock, flags);
+	if (irq->target_vcpu != vcpu) {
+		raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
+		goto retry;
+	}
+	irq->pending_latch = status;
+	vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
+
+	if (status) {
+		/* clear consumed data */
+		val &= ~(1 << bit_nr);
+		ret = kvm_write_guest_lock(kvm, ptr, &val, 1);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+/**
+ * vgic_its_save_pending_tables - Save the pending tables into guest RAM
+ * kvm lock and all vcpu lock must be held
+ */
+int vgic_v3_save_pending_tables(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct vgic_irq *irq;
+	gpa_t last_ptr = ~(gpa_t)0;
+	int ret;
+	u8 val;
+
+	list_for_each_entry(irq, &dist->lpi_list_head, lpi_list) {
+		int byte_offset, bit_nr;
+		struct kvm_vcpu *vcpu;
+		gpa_t pendbase, ptr;
+		bool stored;
+
+		vcpu = irq->target_vcpu;
+		if (!vcpu)
+			continue;
+
+		pendbase = GICR_PENDBASER_ADDRESS(vcpu->arch.vgic_cpu.pendbaser);
+
+		byte_offset = irq->intid / BITS_PER_BYTE;
+		bit_nr = irq->intid % BITS_PER_BYTE;
+		ptr = pendbase + byte_offset;
+
+		if (ptr != last_ptr) {
+			ret = kvm_read_guest_lock(kvm, ptr, &val, 1);
+			if (ret)
+				return ret;
+			last_ptr = ptr;
+		}
+
+		stored = val & (1U << bit_nr);
+		if (stored == irq->pending_latch)
+			continue;
+
+		if (irq->pending_latch)
+			val |= 1 << bit_nr;
+		else
+			val &= ~(1 << bit_nr);
+
+		ret = kvm_write_guest_lock(kvm, ptr, &val, 1);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+/**
+ * vgic_v3_rdist_overlap - check if a region overlaps with any
+ * existing redistributor region
+ *
+ * @kvm: kvm handle
+ * @base: base of the region
+ * @size: size of region
+ *
+ * Return: true if there is an overlap
+ */
+bool vgic_v3_rdist_overlap(struct kvm *kvm, gpa_t base, size_t size)
+{
+	struct vgic_dist *d = &kvm->arch.vgic;
+	struct vgic_redist_region *rdreg;
+
+	list_for_each_entry(rdreg, &d->rd_regions, list) {
+		if ((base + size > rdreg->base) &&
+			(base < rdreg->base + vgic_v3_rd_region_size(kvm, rdreg)))
+			return true;
+	}
+	return false;
+}
+
+/*
+ * Check for overlapping regions and for regions crossing the end of memory
+ * for base addresses which have already been set.
+ */
+bool vgic_v3_check_base(struct kvm *kvm)
+{
+	struct vgic_dist *d = &kvm->arch.vgic;
+	struct vgic_redist_region *rdreg;
+
+	if (!IS_VGIC_ADDR_UNDEF(d->vgic_dist_base) &&
+	    d->vgic_dist_base + KVM_VGIC_V3_DIST_SIZE < d->vgic_dist_base)
+		return false;
+
+	list_for_each_entry(rdreg, &d->rd_regions, list) {
+		if (rdreg->base + vgic_v3_rd_region_size(kvm, rdreg) <
+			rdreg->base)
+			return false;
+	}
+
+	if (IS_VGIC_ADDR_UNDEF(d->vgic_dist_base))
+		return true;
+
+	return !vgic_v3_rdist_overlap(kvm, d->vgic_dist_base,
+				      KVM_VGIC_V3_DIST_SIZE);
+}
+
+/**
+ * vgic_v3_rdist_free_slot - Look up registered rdist regions and identify one
+ * which has free space to put a new rdist region.
+ *
+ * @rd_regions: redistributor region list head
+ *
+ * A redistributor regions maps n redistributors, n = region size / (2 x 64kB).
+ * Stride between redistributors is 0 and regions are filled in the index order.
+ *
+ * Return: the redist region handle, if any, that has space to map a new rdist
+ * region.
+ */
+struct vgic_redist_region *vgic_v3_rdist_free_slot(struct list_head *rd_regions)
+{
+	struct vgic_redist_region *rdreg;
+
+	list_for_each_entry(rdreg, rd_regions, list) {
+		if (!vgic_v3_redist_region_full(rdreg))
+			return rdreg;
+	}
+	return NULL;
+}
+
+struct vgic_redist_region *vgic_v3_rdist_region_from_index(struct kvm *kvm,
+							   u32 index)
+{
+	struct list_head *rd_regions = &kvm->arch.vgic.rd_regions;
+	struct vgic_redist_region *rdreg;
+
+	list_for_each_entry(rdreg, rd_regions, list) {
+		if (rdreg->index == index)
+			return rdreg;
+	}
+	return NULL;
+}
+
+
+int vgic_v3_map_resources(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu;
+	int ret = 0;
+	int c;
+
+	if (vgic_ready(kvm))
+		goto out;
+
+	kvm_for_each_vcpu(c, vcpu, kvm) {
+		struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+		if (IS_VGIC_ADDR_UNDEF(vgic_cpu->rd_iodev.base_addr)) {
+			kvm_debug("vcpu %d redistributor base not set\n", c);
+			ret = -ENXIO;
+			goto out;
+		}
+	}
+
+	if (IS_VGIC_ADDR_UNDEF(dist->vgic_dist_base)) {
+		kvm_err("Need to set vgic distributor addresses first\n");
+		ret = -ENXIO;
+		goto out;
+	}
+
+	if (!vgic_v3_check_base(kvm)) {
+		kvm_err("VGIC redist and dist frames overlap\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/*
+	 * For a VGICv3 we require the userland to explicitly initialize
+	 * the VGIC before we need to use it.
+	 */
+	if (!vgic_initialized(kvm)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	ret = vgic_register_dist_iodev(kvm, dist->vgic_dist_base, VGIC_V3);
+	if (ret) {
+		kvm_err("Unable to register VGICv3 dist MMIO regions\n");
+		goto out;
+	}
+
+	dist->ready = true;
+
+out:
+	return ret;
+}
+
+DEFINE_STATIC_KEY_FALSE(vgic_v3_cpuif_trap);
+
+static int __init early_group0_trap_cfg(char *buf)
+{
+	return strtobool(buf, &group0_trap);
+}
+early_param("kvm-arm.vgic_v3_group0_trap", early_group0_trap_cfg);
+
+static int __init early_group1_trap_cfg(char *buf)
+{
+	return strtobool(buf, &group1_trap);
+}
+early_param("kvm-arm.vgic_v3_group1_trap", early_group1_trap_cfg);
+
+static int __init early_common_trap_cfg(char *buf)
+{
+	return strtobool(buf, &common_trap);
+}
+early_param("kvm-arm.vgic_v3_common_trap", early_common_trap_cfg);
+
+static int __init early_gicv4_enable(char *buf)
+{
+	return strtobool(buf, &gicv4_enable);
+}
+early_param("kvm-arm.vgic_v4_enable", early_gicv4_enable);
+
+/**
+ * vgic_v3_probe - probe for a VGICv3 compatible interrupt controller
+ * @info:	pointer to the GIC description
+ *
+ * Returns 0 if the VGICv3 has been probed successfully, returns an error code
+ * otherwise
+ */
+int vgic_v3_probe(const struct gic_kvm_info *info)
+{
+	u32 ich_vtr_el2 = kvm_call_hyp_ret(__vgic_v3_get_ich_vtr_el2);
+	int ret;
+
+	/*
+	 * The ListRegs field is 5 bits, but there is a architectural
+	 * maximum of 16 list registers. Just ignore bit 4...
+	 */
+	kvm_vgic_global_state.nr_lr = (ich_vtr_el2 & 0xf) + 1;
+	kvm_vgic_global_state.can_emulate_gicv2 = false;
+	kvm_vgic_global_state.ich_vtr_el2 = ich_vtr_el2;
+
+	/* GICv4 support? */
+	if (info->has_v4) {
+		kvm_vgic_global_state.has_gicv4 = gicv4_enable;
+		kvm_info("GICv4 support %sabled\n",
+			 gicv4_enable ? "en" : "dis");
+	}
+
+	if (!info->vcpu.start) {
+		kvm_info("GICv3: no GICV resource entry\n");
+		kvm_vgic_global_state.vcpu_base = 0;
+	} else if (!PAGE_ALIGNED(info->vcpu.start)) {
+		pr_warn("GICV physical address 0x%llx not page aligned\n",
+			(unsigned long long)info->vcpu.start);
+		kvm_vgic_global_state.vcpu_base = 0;
+	} else {
+		kvm_vgic_global_state.vcpu_base = info->vcpu.start;
+		kvm_vgic_global_state.can_emulate_gicv2 = true;
+		ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+		if (ret) {
+			kvm_err("Cannot register GICv2 KVM device.\n");
+			return ret;
+		}
+		kvm_info("vgic-v2@%llx\n", info->vcpu.start);
+	}
+	ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V3);
+	if (ret) {
+		kvm_err("Cannot register GICv3 KVM device.\n");
+		kvm_unregister_device_ops(KVM_DEV_TYPE_ARM_VGIC_V2);
+		return ret;
+	}
+
+	if (kvm_vgic_global_state.vcpu_base == 0)
+		kvm_info("disabling GICv2 emulation\n");
+
+#ifdef CONFIG_ARM64
+	if (cpus_have_const_cap(ARM64_WORKAROUND_CAVIUM_30115)) {
+		group0_trap = true;
+		group1_trap = true;
+	}
+#endif
+
+	if (group0_trap || group1_trap || common_trap) {
+		kvm_info("GICv3 sysreg trapping enabled ([%s%s%s], reduced performance)\n",
+			 group0_trap ? "G0" : "",
+			 group1_trap ? "G1" : "",
+			 common_trap ? "C"  : "");
+		static_branch_enable(&vgic_v3_cpuif_trap);
+	}
+
+	kvm_vgic_global_state.vctrl_base = NULL;
+	kvm_vgic_global_state.type = VGIC_V3;
+	kvm_vgic_global_state.max_gic_vcpus = VGIC_V3_MAX_CPUS;
+
+	return 0;
+}
+
+void vgic_v3_load(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+
+	/*
+	 * If dealing with a GICv2 emulation on GICv3, VMCR_EL2.VFIQen
+	 * is dependent on ICC_SRE_EL1.SRE, and we have to perform the
+	 * VMCR_EL2 save/restore in the world switch.
+	 */
+	if (likely(cpu_if->vgic_sre))
+		kvm_call_hyp(__vgic_v3_write_vmcr, cpu_if->vgic_vmcr);
+
+	kvm_call_hyp(__vgic_v3_restore_aprs, vcpu);
+
+	if (has_vhe())
+		__vgic_v3_activate_traps(vcpu);
+}
+
+void vgic_v3_vmcr_sync(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
+
+	if (likely(cpu_if->vgic_sre))
+		cpu_if->vgic_vmcr = kvm_call_hyp_ret(__vgic_v3_read_vmcr);
+}
+
+void vgic_v3_put(struct kvm_vcpu *vcpu)
+{
+	vgic_v3_vmcr_sync(vcpu);
+
+	kvm_call_hyp(__vgic_v3_save_aprs, vcpu);
+
+	if (has_vhe())
+		__vgic_v3_deactivate_traps(vcpu);
+}

virt/kvm/arm/vgic/vgic-v4.c

@@ -0,0 +1,355 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2017 ARM Ltd.
+ * Author: Marc Zyngier <marc.zyngier@arm.com>
+ */
+
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+#include <linux/kvm_host.h>
+#include <linux/irqchip/arm-gic-v3.h>
+
+#include "vgic.h"
+
+/*
+ * How KVM uses GICv4 (insert rude comments here):
+ *
+ * The vgic-v4 layer acts as a bridge between several entities:
+ * - The GICv4 ITS representation offered by the ITS driver
+ * - VFIO, which is in charge of the PCI endpoint
+ * - The virtual ITS, which is the only thing the guest sees
+ *
+ * The configuration of VLPIs is triggered by a callback from VFIO,
+ * instructing KVM that a PCI device has been configured to deliver
+ * MSIs to a vITS.
+ *
+ * kvm_vgic_v4_set_forwarding() is thus called with the routing entry,
+ * and this is used to find the corresponding vITS data structures
+ * (ITS instance, device, event and irq) using a process that is
+ * extremely similar to the injection of an MSI.
+ *
+ * At this stage, we can link the guest's view of an LPI (uniquely
+ * identified by the routing entry) and the host irq, using the GICv4
+ * driver mapping operation. Should the mapping succeed, we've then
+ * successfully upgraded the guest's LPI to a VLPI. We can then start
+ * with updating GICv4's view of the property table and generating an
+ * INValidation in order to kickstart the delivery of this VLPI to the
+ * guest directly, without software intervention. Well, almost.
+ *
+ * When the PCI endpoint is deconfigured, this operation is reversed
+ * with VFIO calling kvm_vgic_v4_unset_forwarding().
+ *
+ * Once the VLPI has been mapped, it needs to follow any change the
+ * guest performs on its LPI through the vITS. For that, a number of
+ * command handlers have hooks to communicate these changes to the HW:
+ * - Any invalidation triggers a call to its_prop_update_vlpi()
+ * - The INT command results in a irq_set_irqchip_state(), which
+ *   generates an INT on the corresponding VLPI.
+ * - The CLEAR command results in a irq_set_irqchip_state(), which
+ *   generates an CLEAR on the corresponding VLPI.
+ * - DISCARD translates into an unmap, similar to a call to
+ *   kvm_vgic_v4_unset_forwarding().
+ * - MOVI is translated by an update of the existing mapping, changing
+ *   the target vcpu, resulting in a VMOVI being generated.
+ * - MOVALL is translated by a string of mapping updates (similar to
+ *   the handling of MOVI). MOVALL is horrible.
+ *
+ * Note that a DISCARD/MAPTI sequence emitted from the guest without
+ * reprogramming the PCI endpoint after MAPTI does not result in a
+ * VLPI being mapped, as there is no callback from VFIO (the guest
+ * will get the interrupt via the normal SW injection). Fixing this is
+ * not trivial, and requires some horrible messing with the VFIO
+ * internals. Not fun. Don't do that.
+ *
+ * Then there is the scheduling. Each time a vcpu is about to run on a
+ * physical CPU, KVM must tell the corresponding redistributor about
+ * it. And if we've migrated our vcpu from one CPU to another, we must
+ * tell the ITS (so that the messages reach the right redistributor).
+ * This is done in two steps: first issue a irq_set_affinity() on the
+ * irq corresponding to the vcpu, then call its_schedule_vpe(). You
+ * must be in a non-preemptible context. On exit, another call to
+ * its_schedule_vpe() tells the redistributor that we're done with the
+ * vcpu.
+ *
+ * Finally, the doorbell handling: Each vcpu is allocated an interrupt
+ * which will fire each time a VLPI is made pending whilst the vcpu is
+ * not running. Each time the vcpu gets blocked, the doorbell
+ * interrupt gets enabled. When the vcpu is unblocked (for whatever
+ * reason), the doorbell interrupt is disabled.
+ */
+
+#define DB_IRQ_FLAGS	(IRQ_NOAUTOEN | IRQ_DISABLE_UNLAZY | IRQ_NO_BALANCING)
+
+static irqreturn_t vgic_v4_doorbell_handler(int irq, void *info)
+{
+	struct kvm_vcpu *vcpu = info;
+
+	vcpu->arch.vgic_cpu.vgic_v3.its_vpe.pending_last = true;
+	kvm_make_request(KVM_REQ_IRQ_PENDING, vcpu);
+	kvm_vcpu_kick(vcpu);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * vgic_v4_init - Initialize the GICv4 data structures
+ * @kvm:	Pointer to the VM being initialized
+ *
+ * We may be called each time a vITS is created, or when the
+ * vgic is initialized. This relies on kvm->lock to be
+ * held. In both cases, the number of vcpus should now be
+ * fixed.
+ */
+int vgic_v4_init(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct kvm_vcpu *vcpu;
+	int i, nr_vcpus, ret;
+
+	if (!kvm_vgic_global_state.has_gicv4)
+		return 0; /* Nothing to see here... move along. */
+
+	if (dist->its_vm.vpes)
+		return 0;
+
+	nr_vcpus = atomic_read(&kvm->online_vcpus);
+
+	dist->its_vm.vpes = kcalloc(nr_vcpus, sizeof(*dist->its_vm.vpes),
+				    GFP_KERNEL);
+	if (!dist->its_vm.vpes)
+		return -ENOMEM;
+
+	dist->its_vm.nr_vpes = nr_vcpus;
+
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		dist->its_vm.vpes[i] = &vcpu->arch.vgic_cpu.vgic_v3.its_vpe;
+
+	ret = its_alloc_vcpu_irqs(&dist->its_vm);
+	if (ret < 0) {
+		kvm_err("VPE IRQ allocation failure\n");
+		kfree(dist->its_vm.vpes);
+		dist->its_vm.nr_vpes = 0;
+		dist->its_vm.vpes = NULL;
+		return ret;
+	}
+
+	kvm_for_each_vcpu(i, vcpu, kvm) {
+		int irq = dist->its_vm.vpes[i]->irq;
+
+		/*
+		 * Don't automatically enable the doorbell, as we're
+		 * flipping it back and forth when the vcpu gets
+		 * blocked. Also disable the lazy disabling, as the
+		 * doorbell could kick us out of the guest too
+		 * early...
+		 */
+		irq_set_status_flags(irq, DB_IRQ_FLAGS);
+		ret = request_irq(irq, vgic_v4_doorbell_handler,
+				  0, "vcpu", vcpu);
+		if (ret) {
+			kvm_err("failed to allocate vcpu IRQ%d\n", irq);
+			/*
+			 * Trick: adjust the number of vpes so we know
+			 * how many to nuke on teardown...
+			 */
+			dist->its_vm.nr_vpes = i;
+			break;
+		}
+	}
+
+	if (ret)
+		vgic_v4_teardown(kvm);
+
+	return ret;
+}
+
+/**
+ * vgic_v4_teardown - Free the GICv4 data structures
+ * @kvm:	Pointer to the VM being destroyed
+ *
+ * Relies on kvm->lock to be held.
+ */
+void vgic_v4_teardown(struct kvm *kvm)
+{
+	struct its_vm *its_vm = &kvm->arch.vgic.its_vm;
+	int i;
+
+	if (!its_vm->vpes)
+		return;
+
+	for (i = 0; i < its_vm->nr_vpes; i++) {
+		struct kvm_vcpu *vcpu = kvm_get_vcpu(kvm, i);
+		int irq = its_vm->vpes[i]->irq;
+
+		irq_clear_status_flags(irq, DB_IRQ_FLAGS);
+		free_irq(irq, vcpu);
+	}
+
+	its_free_vcpu_irqs(its_vm);
+	kfree(its_vm->vpes);
+	its_vm->nr_vpes = 0;
+	its_vm->vpes = NULL;
+}
+
+int vgic_v4_sync_hwstate(struct kvm_vcpu *vcpu)
+{
+	if (!vgic_supports_direct_msis(vcpu->kvm))
+		return 0;
+
+	return its_schedule_vpe(&vcpu->arch.vgic_cpu.vgic_v3.its_vpe, false);
+}
+
+int vgic_v4_flush_hwstate(struct kvm_vcpu *vcpu)
+{
+	int irq = vcpu->arch.vgic_cpu.vgic_v3.its_vpe.irq;
+	int err;
+
+	if (!vgic_supports_direct_msis(vcpu->kvm))
+		return 0;
+
+	/*
+	 * Before making the VPE resident, make sure the redistributor
+	 * corresponding to our current CPU expects us here. See the
+	 * doc in drivers/irqchip/irq-gic-v4.c to understand how this
+	 * turns into a VMOVP command at the ITS level.
+	 */
+	err = irq_set_affinity(irq, cpumask_of(smp_processor_id()));
+	if (err)
+		return err;
+
+	err = its_schedule_vpe(&vcpu->arch.vgic_cpu.vgic_v3.its_vpe, true);
+	if (err)
+		return err;
+
+	/*
+	 * Now that the VPE is resident, let's get rid of a potential
+	 * doorbell interrupt that would still be pending.
+	 */
+	err = irq_set_irqchip_state(irq, IRQCHIP_STATE_PENDING, false);
+
+	return err;
+}
+
+static struct vgic_its *vgic_get_its(struct kvm *kvm,
+				     struct kvm_kernel_irq_routing_entry *irq_entry)
+{
+	struct kvm_msi msi  = (struct kvm_msi) {
+		.address_lo	= irq_entry->msi.address_lo,
+		.address_hi	= irq_entry->msi.address_hi,
+		.data		= irq_entry->msi.data,
+		.flags		= irq_entry->msi.flags,
+		.devid		= irq_entry->msi.devid,
+	};
+
+	return vgic_msi_to_its(kvm, &msi);
+}
+
+int kvm_vgic_v4_set_forwarding(struct kvm *kvm, int virq,
+			       struct kvm_kernel_irq_routing_entry *irq_entry)
+{
+	struct vgic_its *its;
+	struct vgic_irq *irq;
+	struct its_vlpi_map map;
+	int ret;
+
+	if (!vgic_supports_direct_msis(kvm))
+		return 0;
+
+	/*
+	 * Get the ITS, and escape early on error (not a valid
+	 * doorbell for any of our vITSs).
+	 */
+	its = vgic_get_its(kvm, irq_entry);
+	if (IS_ERR(its))
+		return 0;
+
+	mutex_lock(&its->its_lock);
+
+	/* Perform then actual DevID/EventID -> LPI translation. */
+	ret = vgic_its_resolve_lpi(kvm, its, irq_entry->msi.devid,
+				   irq_entry->msi.data, &irq);
+	if (ret)
+		goto out;
+
+	/*
+	 * Emit the mapping request. If it fails, the ITS probably
+	 * isn't v4 compatible, so let's silently bail out. Holding
+	 * the ITS lock should ensure that nothing can modify the
+	 * target vcpu.
+	 */
+	map = (struct its_vlpi_map) {
+		.vm		= &kvm->arch.vgic.its_vm,
+		.vpe		= &irq->target_vcpu->arch.vgic_cpu.vgic_v3.its_vpe,
+		.vintid		= irq->intid,
+		.properties	= ((irq->priority & 0xfc) |
+				   (irq->enabled ? LPI_PROP_ENABLED : 0) |
+				   LPI_PROP_GROUP1),
+		.db_enabled	= true,
+	};
+
+	ret = its_map_vlpi(virq, &map);
+	if (ret)
+		goto out;
+
+	irq->hw		= true;
+	irq->host_irq	= virq;
+
+out:
+	mutex_unlock(&its->its_lock);
+	return ret;
+}
+
+int kvm_vgic_v4_unset_forwarding(struct kvm *kvm, int virq,
+				 struct kvm_kernel_irq_routing_entry *irq_entry)
+{
+	struct vgic_its *its;
+	struct vgic_irq *irq;
+	int ret;
+
+	if (!vgic_supports_direct_msis(kvm))
+		return 0;
+
+	/*
+	 * Get the ITS, and escape early on error (not a valid
+	 * doorbell for any of our vITSs).
+	 */
+	its = vgic_get_its(kvm, irq_entry);
+	if (IS_ERR(its))
+		return 0;
+
+	mutex_lock(&its->its_lock);
+
+	ret = vgic_its_resolve_lpi(kvm, its, irq_entry->msi.devid,
+				   irq_entry->msi.data, &irq);
+	if (ret)
+		goto out;
+
+	WARN_ON(!(irq->hw && irq->host_irq == virq));
+	if (irq->hw) {
+		irq->hw = false;
+		ret = its_unmap_vlpi(virq);
+	}
+
+out:
+	mutex_unlock(&its->its_lock);
+	return ret;
+}
+
+void kvm_vgic_v4_enable_doorbell(struct kvm_vcpu *vcpu)
+{
+	if (vgic_supports_direct_msis(vcpu->kvm)) {
+		int irq = vcpu->arch.vgic_cpu.vgic_v3.its_vpe.irq;
+		if (irq)
+			enable_irq(irq);
+	}
+}
+
+void kvm_vgic_v4_disable_doorbell(struct kvm_vcpu *vcpu)
+{
+	if (vgic_supports_direct_msis(vcpu->kvm)) {
+		int irq = vcpu->arch.vgic_cpu.vgic_v3.its_vpe.irq;
+		if (irq)
+			disable_irq(irq);
+	}
+}

virt/kvm/arm/vgic/vgic.h

@@ -0,0 +1,322 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ */
+#ifndef __KVM_ARM_VGIC_NEW_H__
+#define __KVM_ARM_VGIC_NEW_H__
+
+#include <linux/irqchip/arm-gic-common.h>
+
+#define PRODUCT_ID_KVM		0x4b	/* ASCII code K */
+#define IMPLEMENTER_ARM		0x43b
+
+#define VGIC_ADDR_UNDEF		(-1)
+#define IS_VGIC_ADDR_UNDEF(_x)  ((_x) == VGIC_ADDR_UNDEF)
+
+#define INTERRUPT_ID_BITS_SPIS	10
+#define INTERRUPT_ID_BITS_ITS	16
+#define VGIC_PRI_BITS		5
+
+#define vgic_irq_is_sgi(intid) ((intid) < VGIC_NR_SGIS)
+
+#define VGIC_AFFINITY_0_SHIFT 0
+#define VGIC_AFFINITY_0_MASK (0xffUL << VGIC_AFFINITY_0_SHIFT)
+#define VGIC_AFFINITY_1_SHIFT 8
+#define VGIC_AFFINITY_1_MASK (0xffUL << VGIC_AFFINITY_1_SHIFT)
+#define VGIC_AFFINITY_2_SHIFT 16
+#define VGIC_AFFINITY_2_MASK (0xffUL << VGIC_AFFINITY_2_SHIFT)
+#define VGIC_AFFINITY_3_SHIFT 24
+#define VGIC_AFFINITY_3_MASK (0xffUL << VGIC_AFFINITY_3_SHIFT)
+
+#define VGIC_AFFINITY_LEVEL(reg, level) \
+	((((reg) & VGIC_AFFINITY_## level ##_MASK) \
+	>> VGIC_AFFINITY_## level ##_SHIFT) << MPIDR_LEVEL_SHIFT(level))
+
+/*
+ * The Userspace encodes the affinity differently from the MPIDR,
+ * Below macro converts vgic userspace format to MPIDR reg format.
+ */
+#define VGIC_TO_MPIDR(val) (VGIC_AFFINITY_LEVEL(val, 0) | \
+			    VGIC_AFFINITY_LEVEL(val, 1) | \
+			    VGIC_AFFINITY_LEVEL(val, 2) | \
+			    VGIC_AFFINITY_LEVEL(val, 3))
+
+/*
+ * As per Documentation/virt/kvm/devices/arm-vgic-v3.txt,
+ * below macros are defined for CPUREG encoding.
+ */
+#define KVM_REG_ARM_VGIC_SYSREG_OP0_MASK   0x000000000000c000
+#define KVM_REG_ARM_VGIC_SYSREG_OP0_SHIFT  14
+#define KVM_REG_ARM_VGIC_SYSREG_OP1_MASK   0x0000000000003800
+#define KVM_REG_ARM_VGIC_SYSREG_OP1_SHIFT  11
+#define KVM_REG_ARM_VGIC_SYSREG_CRN_MASK   0x0000000000000780
+#define KVM_REG_ARM_VGIC_SYSREG_CRN_SHIFT  7
+#define KVM_REG_ARM_VGIC_SYSREG_CRM_MASK   0x0000000000000078
+#define KVM_REG_ARM_VGIC_SYSREG_CRM_SHIFT  3
+#define KVM_REG_ARM_VGIC_SYSREG_OP2_MASK   0x0000000000000007
+#define KVM_REG_ARM_VGIC_SYSREG_OP2_SHIFT  0
+
+#define KVM_DEV_ARM_VGIC_SYSREG_MASK (KVM_REG_ARM_VGIC_SYSREG_OP0_MASK | \
+				      KVM_REG_ARM_VGIC_SYSREG_OP1_MASK | \
+				      KVM_REG_ARM_VGIC_SYSREG_CRN_MASK | \
+				      KVM_REG_ARM_VGIC_SYSREG_CRM_MASK | \
+				      KVM_REG_ARM_VGIC_SYSREG_OP2_MASK)
+
+/*
+ * As per Documentation/virt/kvm/devices/arm-vgic-its.txt,
+ * below macros are defined for ITS table entry encoding.
+ */
+#define KVM_ITS_CTE_VALID_SHIFT		63
+#define KVM_ITS_CTE_VALID_MASK		BIT_ULL(63)
+#define KVM_ITS_CTE_RDBASE_SHIFT	16
+#define KVM_ITS_CTE_ICID_MASK		GENMASK_ULL(15, 0)
+#define KVM_ITS_ITE_NEXT_SHIFT		48
+#define KVM_ITS_ITE_PINTID_SHIFT	16
+#define KVM_ITS_ITE_PINTID_MASK		GENMASK_ULL(47, 16)
+#define KVM_ITS_ITE_ICID_MASK		GENMASK_ULL(15, 0)
+#define KVM_ITS_DTE_VALID_SHIFT		63
+#define KVM_ITS_DTE_VALID_MASK		BIT_ULL(63)
+#define KVM_ITS_DTE_NEXT_SHIFT		49
+#define KVM_ITS_DTE_NEXT_MASK		GENMASK_ULL(62, 49)
+#define KVM_ITS_DTE_ITTADDR_SHIFT	5
+#define KVM_ITS_DTE_ITTADDR_MASK	GENMASK_ULL(48, 5)
+#define KVM_ITS_DTE_SIZE_MASK		GENMASK_ULL(4, 0)
+#define KVM_ITS_L1E_VALID_MASK		BIT_ULL(63)
+/* we only support 64 kB translation table page size */
+#define KVM_ITS_L1E_ADDR_MASK		GENMASK_ULL(51, 16)
+
+#define KVM_VGIC_V3_RDIST_INDEX_MASK	GENMASK_ULL(11, 0)
+#define KVM_VGIC_V3_RDIST_FLAGS_MASK	GENMASK_ULL(15, 12)
+#define KVM_VGIC_V3_RDIST_FLAGS_SHIFT	12
+#define KVM_VGIC_V3_RDIST_BASE_MASK	GENMASK_ULL(51, 16)
+#define KVM_VGIC_V3_RDIST_COUNT_MASK	GENMASK_ULL(63, 52)
+#define KVM_VGIC_V3_RDIST_COUNT_SHIFT	52
+
+#ifdef CONFIG_DEBUG_SPINLOCK
+#define DEBUG_SPINLOCK_BUG_ON(p) BUG_ON(p)
+#else
+#define DEBUG_SPINLOCK_BUG_ON(p)
+#endif
+
+/* Requires the irq_lock to be held by the caller. */
+static inline bool irq_is_pending(struct vgic_irq *irq)
+{
+	if (irq->config == VGIC_CONFIG_EDGE)
+		return irq->pending_latch;
+	else
+		return irq->pending_latch || irq->line_level;
+}
+
+static inline bool vgic_irq_is_mapped_level(struct vgic_irq *irq)
+{
+	return irq->config == VGIC_CONFIG_LEVEL && irq->hw;
+}
+
+static inline int vgic_irq_get_lr_count(struct vgic_irq *irq)
+{
+	/* Account for the active state as an interrupt */
+	if (vgic_irq_is_sgi(irq->intid) && irq->source)
+		return hweight8(irq->source) + irq->active;
+
+	return irq_is_pending(irq) || irq->active;
+}
+
+static inline bool vgic_irq_is_multi_sgi(struct vgic_irq *irq)
+{
+	return vgic_irq_get_lr_count(irq) > 1;
+}
+
+/*
+ * This struct provides an intermediate representation of the fields contained
+ * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC
+ * state to userspace can generate either GICv2 or GICv3 CPU interface
+ * registers regardless of the hardware backed GIC used.
+ */
+struct vgic_vmcr {
+	u32	grpen0;
+	u32	grpen1;
+
+	u32	ackctl;
+	u32	fiqen;
+	u32	cbpr;
+	u32	eoim;
+
+	u32	abpr;
+	u32	bpr;
+	u32	pmr;  /* Priority mask field in the GICC_PMR and
+		       * ICC_PMR_EL1 priority field format */
+};
+
+struct vgic_reg_attr {
+	struct kvm_vcpu *vcpu;
+	gpa_t addr;
+};
+
+int vgic_v3_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr,
+		       struct vgic_reg_attr *reg_attr);
+int vgic_v2_parse_attr(struct kvm_device *dev, struct kvm_device_attr *attr,
+		       struct vgic_reg_attr *reg_attr);
+const struct vgic_register_region *
+vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev,
+		     gpa_t addr, int len);
+struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
+			      u32 intid);
+void __vgic_put_lpi_locked(struct kvm *kvm, struct vgic_irq *irq);
+void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq);
+bool vgic_get_phys_line_level(struct vgic_irq *irq);
+void vgic_irq_set_phys_pending(struct vgic_irq *irq, bool pending);
+void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active);
+bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
+			   unsigned long flags);
+void vgic_kick_vcpus(struct kvm *kvm);
+
+int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
+		      phys_addr_t addr, phys_addr_t alignment);
+
+void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu);
+void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr);
+void vgic_v2_clear_lr(struct kvm_vcpu *vcpu, int lr);
+void vgic_v2_set_underflow(struct kvm_vcpu *vcpu);
+void vgic_v2_set_npie(struct kvm_vcpu *vcpu);
+int vgic_v2_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr);
+int vgic_v2_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 int offset, u32 *val);
+int vgic_v2_cpuif_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			  int offset, u32 *val);
+void vgic_v2_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+void vgic_v2_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+void vgic_v2_enable(struct kvm_vcpu *vcpu);
+int vgic_v2_probe(const struct gic_kvm_info *info);
+int vgic_v2_map_resources(struct kvm *kvm);
+int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
+			     enum vgic_type);
+
+void vgic_v2_init_lrs(void);
+void vgic_v2_load(struct kvm_vcpu *vcpu);
+void vgic_v2_put(struct kvm_vcpu *vcpu);
+void vgic_v2_vmcr_sync(struct kvm_vcpu *vcpu);
+
+void vgic_v2_save_state(struct kvm_vcpu *vcpu);
+void vgic_v2_restore_state(struct kvm_vcpu *vcpu);
+
+static inline void vgic_get_irq_kref(struct vgic_irq *irq)
+{
+	if (irq->intid < VGIC_MIN_LPI)
+		return;
+
+	kref_get(&irq->refcount);
+}
+
+void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu);
+void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr);
+void vgic_v3_clear_lr(struct kvm_vcpu *vcpu, int lr);
+void vgic_v3_set_underflow(struct kvm_vcpu *vcpu);
+void vgic_v3_set_npie(struct kvm_vcpu *vcpu);
+void vgic_v3_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+void vgic_v3_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+void vgic_v3_enable(struct kvm_vcpu *vcpu);
+int vgic_v3_probe(const struct gic_kvm_info *info);
+int vgic_v3_map_resources(struct kvm *kvm);
+int vgic_v3_lpi_sync_pending_status(struct kvm *kvm, struct vgic_irq *irq);
+int vgic_v3_save_pending_tables(struct kvm *kvm);
+int vgic_v3_set_redist_base(struct kvm *kvm, u32 index, u64 addr, u32 count);
+int vgic_register_redist_iodev(struct kvm_vcpu *vcpu);
+bool vgic_v3_check_base(struct kvm *kvm);
+
+void vgic_v3_load(struct kvm_vcpu *vcpu);
+void vgic_v3_put(struct kvm_vcpu *vcpu);
+void vgic_v3_vmcr_sync(struct kvm_vcpu *vcpu);
+
+bool vgic_has_its(struct kvm *kvm);
+int kvm_vgic_register_its_device(void);
+void vgic_enable_lpis(struct kvm_vcpu *vcpu);
+void vgic_flush_pending_lpis(struct kvm_vcpu *vcpu);
+int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi);
+int vgic_v3_has_attr_regs(struct kvm_device *dev, struct kvm_device_attr *attr);
+int vgic_v3_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 int offset, u32 *val);
+int vgic_v3_redist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 int offset, u32 *val);
+int vgic_v3_cpu_sysregs_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+			 u64 id, u64 *val);
+int vgic_v3_has_cpu_sysregs_attr(struct kvm_vcpu *vcpu, bool is_write, u64 id,
+				u64 *reg);
+int vgic_v3_line_level_info_uaccess(struct kvm_vcpu *vcpu, bool is_write,
+				    u32 intid, u64 *val);
+int kvm_register_vgic_device(unsigned long type);
+void vgic_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+void vgic_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
+int vgic_lazy_init(struct kvm *kvm);
+int vgic_init(struct kvm *kvm);
+
+void vgic_debug_init(struct kvm *kvm);
+void vgic_debug_destroy(struct kvm *kvm);
+
+bool lock_all_vcpus(struct kvm *kvm);
+void unlock_all_vcpus(struct kvm *kvm);
+
+static inline int vgic_v3_max_apr_idx(struct kvm_vcpu *vcpu)
+{
+	struct vgic_cpu *cpu_if = &vcpu->arch.vgic_cpu;
+
+	/*
+	 * num_pri_bits are initialized with HW supported values.
+	 * We can rely safely on num_pri_bits even if VM has not
+	 * restored ICC_CTLR_EL1 before restoring APnR registers.
+	 */
+	switch (cpu_if->num_pri_bits) {
+	case 7: return 3;
+	case 6: return 1;
+	default: return 0;
+	}
+}
+
+static inline bool
+vgic_v3_redist_region_full(struct vgic_redist_region *region)
+{
+	if (!region->count)
+		return false;
+
+	return (region->free_index >= region->count);
+}
+
+struct vgic_redist_region *vgic_v3_rdist_free_slot(struct list_head *rdregs);
+
+static inline size_t
+vgic_v3_rd_region_size(struct kvm *kvm, struct vgic_redist_region *rdreg)
+{
+	if (!rdreg->count)
+		return atomic_read(&kvm->online_vcpus) * KVM_VGIC_V3_REDIST_SIZE;
+	else
+		return rdreg->count * KVM_VGIC_V3_REDIST_SIZE;
+}
+
+struct vgic_redist_region *vgic_v3_rdist_region_from_index(struct kvm *kvm,
+							   u32 index);
+
+bool vgic_v3_rdist_overlap(struct kvm *kvm, gpa_t base, size_t size);
+
+static inline bool vgic_dist_overlap(struct kvm *kvm, gpa_t base, size_t size)
+{
+	struct vgic_dist *d = &kvm->arch.vgic;
+
+	return (base + size > d->vgic_dist_base) &&
+		(base < d->vgic_dist_base + KVM_VGIC_V3_DIST_SIZE);
+}
+
+int vgic_copy_lpi_list(struct kvm *kvm, struct kvm_vcpu *vcpu, u32 **intid_ptr);
+int vgic_its_resolve_lpi(struct kvm *kvm, struct vgic_its *its,
+			 u32 devid, u32 eventid, struct vgic_irq **irq);
+struct vgic_its *vgic_msi_to_its(struct kvm *kvm, struct kvm_msi *msi);
+int vgic_its_inject_cached_translation(struct kvm *kvm, struct kvm_msi *msi);
+void vgic_lpi_translation_cache_init(struct kvm *kvm);
+void vgic_lpi_translation_cache_destroy(struct kvm *kvm);
+void vgic_its_invalidate_cache(struct kvm *kvm);
+
+bool vgic_supports_direct_msis(struct kvm *kvm);
+int vgic_v4_init(struct kvm *kvm);
+void vgic_v4_teardown(struct kvm *kvm);
+int vgic_v4_sync_hwstate(struct kvm_vcpu *vcpu);
+int vgic_v4_flush_hwstate(struct kvm_vcpu *vcpu);
+
+#endif